Cleveland Clinic Journal of Medicine

Obesity is a leading public health concern, affecting nearly 60 million adult Americans.¹ It is a major risk factor for the development of insulin resistance and type 2 diabetes mellitus (DM).² More than 90% of patients with type 2 DM have obesity, and obesity is a major obstacle to achieving long-term glycemic control.³

Clinical studies have demonstrated that a 6- to 7-kg increase in body weight increases the risk of developing type 2 DM by 50%, while a 5-kg loss reduces the risk by a similar amount.⁴ As a result, most patients who have a body mass index greater than 40 kg/m² suffer from type 2 DM.⁵ Strong evidence exists that bariatric surgery and its resulting weight loss has positive effects on fasting blood sugar, hemoglobin A1c (HbA1c), lipid profiles, and other metabolic variables.⁶

When combined, obesity and type 2 DM carry a significant burden of micro- and macrovascular complications such as retinopathy, nephropathy, neuropathy, and cardiovascular disease. As a result, a high prevalence of morbidity and mortality is seen among patients with obesity and type 2 DM; those between the ages of 51 and 61 have a 7-times higher mortality rate compared with nonobese normoglycemic people, and patients with diabetes alone have a 2.6-times higher mortality rate.⁷

A DILEMMA IN THE CLINIC: FOCUS ON THE SUGAR OR THE WEIGHT?

Although type 2 DM and obesity go hand in hand, clinicians tend to focus on the sugar and neglect the weight, concentrating their efforts on improving blood glucose indices, and prescribing in many instances medications that cause weight gain. As a result, we are faced with a rising epidemic of obesity, perpetuating a preexisting epidemic of diabetes. 

An optimal, comprehensive approach to managing patients with type 2 DM should encompass both the control of dysglycemia and its associated comorbidities, obesity being the key player.⁸ However, clinical  practice is often misaligned with the evidence. For instance, many of our first-line oral treatments for type 2 DM (except for metformin) are associated with weight gain.⁹ With time, control of glycemia becomes more and more ineffective, at which point therapy is intensified with insulin, further exacerbating the weight gain.¹⁰

Therefore, it seems counterintuitive to treat a disease for which obesity is one of the main risk factors with medications that promote weight gain. Yet healthcare providers are faced with a therapeutic dilemma: should they focus their efforts on improving patients’ glycemic control, or should they invest in helping these patients lose weight? Although an ideal approach would incorporate both aspects, the reality is that it is far from practical.

A few issues impinge on integrating weight loss in the care of type 2 DM. Although the American Medical Association recognized obesity as a disease in 2013,¹¹ some providers still perceive obesity as a self-inflicted condition that is due to bad lifestyle and behavior.¹¹ Many clinicians may also have low expectations for patients’ success, and often lack the time and knowledge to intervene regarding nutrition, physical activity, and psychological issues pertinent to the management of obesity in type 2 DM. Therefore, in many cases, it seems less complicated and more rewarding for both patients and physicians to concentrate on improving the HbA1c value rather than investing efforts in weight loss. For diabetic patients with obesity, this could mean that clinicians may prescribe glucose-lowering therapies, such as insulin and sulfonylureas, at the expense of weight gain. Additionally, clinicians often experience the need to provide recommendations more aligned with metrics that dictate reimbursement (eg, HbA1c targets) within healthcare systems that still raise concerns regarding obesity visit reimbursements.

Lastly, the lack of trustworthy or pertinent evidence (lack of comparative effectiveness research) for antiobesity medications may limit their use in daily practice. Physicians have had little confidence in the efficacy of antiobesity drugs, and often raise significant safety concerns, especially after witnessing important fiascos in this field, eg, dexfenfluramine, rimonabant, and sibutramine.^2,12,13

As a result, many of our patients with obesity and type 2 DM may not consider the need for weight loss, and may not even be aware that type 2 DM is caused by obesity and physical inactivity in the first place. Others have accumulated a significant degree of frustration, and have “thrown in the towel” already after unsuccessful weight-loss efforts, many of which were not medically supervised.

For all of the above reasons, both clinicians and patients often concentrate their efforts on treating blood glucose numbers rather than the “obesity-diabetes” as a whole.¹⁴ And as a result, our practices are slowly filling up with patients with obesity and type 2 DM who are treated primarily with insulin, resulting in a progressive (and untreated) obesity and diabetes epidemic.

DRUGS FOR TREATING OBESITY AND TYPE 2 DM

Orlistat

Orlistat (Xenical) is the only weight-loss drug approved by the US Food and Drug Administration (FDA) that acts outside the brain. It inhibits pancreatic lipases, resulting in up to 30% less fat absorption in the gut. Orlistat has been approved for long-term use by the FDA.

Benefits. In the XENical in the Prevention of Diabetes in Obese Subjects study, treatment with orlistat resulted in a significant reduction in the cumulative incidence of type 2 DM after 4 years of treatment (9.0% with placebo vs 6.2% with orlistat), corresponding to a risk reduction of 37.3%.¹⁶ Mean weight loss after 4 years was significantly greater in the orlistat group (5.8 vs 3.0 kg with placebo; P < .001).¹⁶ Other benefits of orlistat included a reduction in low-density lipoprotein cholesterol independent of that expected from change in body weight.¹⁶

Adverse effects include flatulence with discharge and fecal urgency after high-fat dietary indiscretions. Serum levels of fat-soluble vitamins (A, D, E, and K) were lower with orlistat than with placebo,¹⁶ and a fat-soluble vitamin supplement should be taken 2 hours before or after taking orlistat. Serious but very uncommon adverse events such as kidney damage have been reported.¹⁷ Kidney and liver function should be monitored while taking orlistat.

Phentermine (Adipex-P, Lomaira), a sympathomimetic amine, is the most commonly prescribed antiobesity drug in the United States. A schedule IV controlled substance, it is FDA-approved for short-term use (up to 12 weeks). Its primary mechanism of action is mediated by reduction in hunger perception. It was first developed in the 1970s and is available in doses ranging from 8 mg to 37.5 mg daily.¹⁸

Benefits. In a randomized trial, at 28 weeks, weight loss was 1.5 kg with placebo and 5.3 kg with phentermine.¹⁹ No long-term (> 1 year) randomized controlled trials of the effectiveness of phentermine monotherapy in weight loss have been conducted.

Adverse effects. Dizziness, dry mouth, insomnia, constipation, and increase in heart rate were most common.¹⁹

Phentermine is contraindicated in patients with coronary artery disease, congestive heart failure, stroke, and uncontrolled hypertension. Currently, no data exist on the long-term cardiovascular effects of phentermine. We believe phentermine, used in patients at low to intermediate cardiovascular risk, is a useful “jumpstart” tool, in combination with lifestyle changes, to achieve weight loss and improve metabolic values for those with type 2 DM and obesity.

Phentermine is a controlled substance per Ohio law. Patients must be seen once a month by the prescribing provider and prescriptions are limited to a 30-day supply, which must be filled within 7 days of the date of the prescription. Phentermine can only be prescribed for a maximum of 3 months and must be discontinued for 6 months before patients are eligible for a new prescription.

Phentermine and topiramate extended-release

Obesity is a product of complex interactions between several neurohormonal pathways. Approaches simultaneously targeting more than one regulatory pathway have become popular and quite efficient strategies in treating patients with obesity.²⁰ Stemming from such approaches, antiobesity drug combinations such as phentermine and topiramate extended-release (Qsymia) have become increasingly recognized and used in clinical practice. The combination of these 2 medications has been approved for long-term use by the FDA.

Phentermine and topiramate extended-release is a fixed-dose combination that was approved for weight loss in 2012. Topiramate, an anticonvulsant, and phentermine exert their anorexigenic effects through regulating various brain neurotransmitters and result in more weight loss when used together than when either is used alone. Several clinical trials evaluated the efficacy of low doses of this combination in weight loss.

Benefits. In a randomized trial in patients with obesity and cardiometabolic diseases, at 56 weeks, the mean weight loss was:

• 1.2% in the placebo group
• 7.8% in the group receiving phentermine 7.5 mg and topiramate 46 mg
• 9.8% in the group receiving phentermine 15 mg and topiramate 92 mg.²¹

Patients in the active treatment groups also had significant improvements in cardiovascular and metabolic risk factors such as waist circumference, systolic blood pressure, and total cholesterol/high-density lipoprotein cholesterol ratio. At 56 weeks, patients with diabetes and prediabetes taking this preparation had greater reductions in HbA1c values, and fewer prediabetes patients progressed to type 2 DM.²¹

Adverse effects most commonly seen were dry mouth, paresthesia, and constipation.²¹

This combination is contraindicated in pregnancy, patients with recent stroke, uncontrolled hypertension, coronary artery disease, glaucoma, hyperthyroidism, or in patients taking monoamine oxidase inhibitors. Women of childbearing age should be tested for pregnancy before starting therapy, and monthly thereafter, and also be advised to use effective methods of contraception while taking the medication. Topiramate has been associated with the development of renal stones and thus should be used with caution in patients with a history of kidney stones.

Bupropion and naltrexone sustained-release

Bupropion and naltrexone sustained-release (Contrave) is another FDA-approved combination drug for chronic weight management. Bupropion is a dopamine and norepinephrine reuptake inhibitor approved for depression and smoking cessation, and naltrexone is an opioid receptor antagonist approved for treating alcohol and opioid dependence. The combination of these 2 medications has been approved for long-term use by the FDA.

Benefits. In a randomized trial in patients with obesity and type 2 DM, weight loss at 56 weeks was:

• 1.8% with placebo
• 5.0% with naltrexone 32 mg and bupropion 360 mg daily.

Absolute reductions in HbA1c were:

• 0.1% with placebo
• 0.6% with naltrexone-bupropion.

Improvements were also seen in other cardiometabolic risk factors such as triglyceride and high-density lipoprotein cholesterol levels.²²

Adverse effects. The most common adverse effect leading to drug discontinuation was nausea. Other adverse effects reported were constipation, headache, vomiting, and dizziness.²²

Naltrexone-bupropion is contraindicated in patients with a history of seizure disorder or a diagnosis of anorexia nervosa or bulimia, or who are on chronic opioid therapy.

Diethylpropion

Diethylpropion (Tenuate, Tenuate Dospan) is a central nervous system stimulant similar to bupropion in its structure. It was approved by the FDA for treating obesity in 1959. It should be used as part of a short-term weight-loss plan, along with a low-calorie diet. Diethylpropion is also a controlled substance and, as with phentermine therapy, patients are required to be seen once a month by their prescriber. Diethylpropion cannot be prescribed for more than 3 months.

Benefits. Weight loss in a randomized trial at 6 months:

• 3.2% with placebo
• 9.8% with diethylpropion 50 mg twice a day.²³

After 6 months, all participants received diethylpropion in an open-label extension for an additional 6 months. At 12 months, the mean weight loss produced by diethylpropion was 10.6%.²³ No differences in heart rate, blood pressure, electrocardiographic results, or psychiatric evaluations were observed.

Adverse effects. As with phentermine, common side effects of diethylpropion include insomnia, dry mouth, dizziness, headache, mild increases in blood pressure, and palpitations.²³

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA for chronic weight management in June 2012. It exerts its effects through binding selectively to central 5-HT_2C serotonin receptors, with poor affinity for 5-HT_2A and 5-HT_2B receptors. Nonselective serotoninergic agents, including fenfluramine and dexfenfluramine, were withdrawn from the market in 1997 after being reported to be associated with valvular heart abnormalities.²⁴ Lorcaserin has been approved for long-term use by the FDA.

Benefits. Mean weight loss at 1 year in the Behavioral Modification and Lorcaserin for Overweight and Obesity Management in Diabetes Mellitus trial²⁵ was:

• 1.5% with placebo
• 5.0% with lorcaserin 10 mg once daily
• 4.5% with lorcaserin 10 mg twice daily.

Absolute reductions in HbA1c values were:

• 0.4% with placebo
• 0.9% with lorcaserin 10 mg once daily
• 1.0% with lorcaserin 10 mg twice daily.

Absolute reductions in fasting plasma glucose values were:

• 11.9 mg/dL with placebo
• 27.4 mg/dL with lorcaserin 10 mg once daily
• 28.4 mg/dL with lorcaserin 10 mg twice daily.²⁵

Adverse effects. The most common adverse effects were headache, dizziness, and fatigue. There was no significant increase in valvulopathy on echocardiography of participants receiving lorcaserin compared with placebo.²⁵

Liraglutide (Saxenda, Victoza) is a glucagon-like peptide-1 (GLP-1) receptor agonist. Native GLP-1 is a hormone secreted by intestinal L cells in response to consumption of fat and carbohydrate-rich foods. It stimulates the release of insulin and suppresses any inappropriately elevated postprandial glucagon levels. In addition to its effect on glucose metabolism, GLP-1 also reduces appetite and delays gastric emptying in humans.²⁶ Unlike the extremely short half-life of native GLP-1 (estimated at 1 to 2 minutes), liraglutide has a half-life of 13 hours, allowing it to be given once daily.²⁶ Liraglutide medication has been approved for long-term use by the FDA.

Benefits. The Liraglutide Effect and Action in Diabetes 1–5 studies compared the effects of liraglutide monotherapy with antidiabetic oral medications or insulin, as well as in combination with antidiabetic oral agents. Liraglutide (Victoza) at doses approved for type 2 DM of 1.2 mg and 1.8 mg daily had significant effects in reducing HbA1c by 0.48% to 1.84% and weight by 2.5 kg to 4 kg.^27,28 At a dose of 3.0 mg, liraglutide (Saxenda) is approved for chronic weight management. This dose of liraglutide has been shown to be effective and safe in patients with type 2 DM and obesity.

In the 56-week SCALE Diabetes trial,²⁹ liraglutide at a dose of 3.0 mg resulted in 6.0% weight reduction, compared with 2.0% in the placebo group. Of participants receiving 3.0 mg of liraglutide, 54.3% achieved more than 5% weight loss at 56 weeks compared with 21.4% with placebo. Liraglutide also resulted in significant improvements in HbA1c (mean change −1.3% vs −0.3% with placebo), fasting and postprandial glucose levels, and fasting glucagon levels.²⁹

The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results trial has shown liraglutide to significantly reduce rates of major cardiovascular events (first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in patients with elevated cardiovascular risk factors.³⁰ These findings make liraglutide a favorable choice for high-risk patients with type 2 DM, obesity, and cardiovascular disease.

It is important to indicate that if a 5% weight loss is not achieved by 3 months with any of these weight-loss medications, it would be reasonable to stop the medication and consider switching to a different medication. These medications work best when combined with diet and increased physical activity. Weight-loss medications should never be used during pregnancy.

Women of childbearing age should be advised to use effective contraception methods while taking any of the above antiobesity medications.

Diabetes medications associated with weight loss: Metformin and SGLT-2 inhibitors

Although not FDA-approved for weight management, metformin has anorexigenic effects that aid in weight loss. It also inhibits hepatic glucose production and improves peripheral insulin sensitivity, making it a useful agent in patients with type 2 DM and obesity.

A meta-analysis of 31 trials showed that metformin reduced body mass index by 5.3% compared with placebo.³¹ Metformin should be considered as a first-line agent in obese patients with type 2 DM.

In healthy people, nearly all glucose is filtered in the glomerulus, but then 98% of it is reabsorbed in the proximal tubule by sodium-glucose cotransporter-2 (SGLT-2). Drugs that inhibit SGLT-2 increase urinary glucose excretion and, as a result, help control hyperglycemia. Another, off-label effect of excreting more glucose is weight loss: a sustained weight loss of about 3 kg to 5 kg in clinical studies.³² Although they can be used as monotherapies, SGLT-2 inhibitors are usually used as add-on therapies in patients with type 2 DM.

AN ALGORITHM FOR TREATMENT

First, we believe that lifestyle interventions by optimization of nutrition and physical activity should be the cornerstone therapy in the management plan of any patient with type 2 DM and obesity. These interventions are best implemented through a comprehensive, multidisciplinary approach that integrates the care of dietitians, physical therapists, exercise physiologists, psychologists, and social workers.³³ Patients need also to be seen frequently, ie, at least once every 3 months. The possibility of seeing patients in group-shared medical appointments on a monthly basis could also be considered. 

We also believe that metformin should be added early in the course of treatment for its known benefits of improving insulin sensitivity and suppressing appetite. Target HbA1c goals and body weight in patients with type 2 diabetes and obesity should be tailored to the individual based on age, general health status, risk of hypoglycemia, capacity to do physical activity, and associated comorbidities. If no improvements are seen (HbA1c > 7% and < 3 % weight loss) despite lifestyle changes and the addition of metformin, the possibility of adding a GLP-1 receptor agonist or an SGLT-2 inhibitor as a second-line therapy should be considered. Both classes of medications aid in lowering HbA1c and promote further weight loss.

If no clinical progress is achieved at 3 months, the possibility of adding an FDA-approved weight-loss medication, as discussed above, should be strongly considered. Of note, this algorithm targets different endogenous pathways for weight loss and thus minimizes weight regain through compensatory mechanisms.

Patient-centered care has become a core quality measure in our healthcare systems and a key to our patients’ success. The decision to start an antiobesity drug should therefore reflect careful consideration of medical and personal patient issues, all of which are valued differently by patients.³⁴

Individualized therapy is even more relevant among patients suffering from a significant burden of disease. About 80% of patients with diabetes live with at least 1 other medical condition,³⁵ and each of these patients spends over 2 hours a day, on average, following doctors’ recommendations.³⁶ If antiobesity medications are prescribed without careful consideration of the patient’s preexisting workload, they will be destined to fail. Therefore, it becomes crucial to first account for the patient’s ability to cope with therapy intensification. This requires careful deliberation between healthcare providers and patients, in aims of targeting a weight-loss plan that fits patients’ goals and is aligned with providers’ expectations.

Healthcare systems also play a key role in supporting better conversations about obesity in type 2 DM patients. They could implement multifaceted initiatives to promote shared decision-making and the use of decision aids to advance patient-centered obesity practices.³⁷ Policymakers could redesign quality measures aimed at capturing the quality of obesity conversations, and develop policies that support better education for clinicians regarding the importance of addressing obesity with adequate communication and patient-centered skills. Guidelines are often too disease-specific and do not consider comorbidities in their context when providing recommendations.³⁸ Thus, diabetes societies should respond to the need to guide care for patients with diabetes and its comorbidities, particularly obesity.

CONCLUSIONS

Obesity is a serious global health issue and a leading risk factor for type 2 DM. Lifestyle measures are the cornerstone of preventing and treating obesity and type 2 DM. Emerging data support the effectiveness of intensive, interdisciplinary weight-loss programs in patients with diabetes. The use of antiobesity drugs should be considered in patients who have not achieved adequate responses to lifestyle interventions. Medications should be tailored to the individual’s health risks and metabolic and psychobehavioral characteristics. In many cases, the addition of weight-loss drugs will help accomplish and maintain the recommended 10% weight reduction, resulting in improvement in glycemic control and significant reduction in cardiovascular risk factors. New studies combining antiobesity and antidiabetes medications in the context of lifestyle interventions will help define the optimal therapeutic approach for patients with type 2 DM and obesity.

Obesity is a leading public health concern, affecting nearly 60 million adult Americans.¹ It is a major risk factor for the development of insulin resistance and type 2 diabetes mellitus (DM).² More than 90% of patients with type 2 DM have obesity, and obesity is a major obstacle to achieving long-term glycemic control.³

Clinical studies have demonstrated that a 6- to 7-kg increase in body weight increases the risk of developing type 2 DM by 50%, while a 5-kg loss reduces the risk by a similar amount.⁴ As a result, most patients who have a body mass index greater than 40 kg/m² suffer from type 2 DM.⁵ Strong evidence exists that bariatric surgery and its resulting weight loss has positive effects on fasting blood sugar, hemoglobin A1c (HbA1c), lipid profiles, and other metabolic variables.⁶

When combined, obesity and type 2 DM carry a significant burden of micro- and macrovascular complications such as retinopathy, nephropathy, neuropathy, and cardiovascular disease. As a result, a high prevalence of morbidity and mortality is seen among patients with obesity and type 2 DM; those between the ages of 51 and 61 have a 7-times higher mortality rate compared with nonobese normoglycemic people, and patients with diabetes alone have a 2.6-times higher mortality rate.⁷

A DILEMMA IN THE CLINIC: FOCUS ON THE SUGAR OR THE WEIGHT?

Although type 2 DM and obesity go hand in hand, clinicians tend to focus on the sugar and neglect the weight, concentrating their efforts on improving blood glucose indices, and prescribing in many instances medications that cause weight gain. As a result, we are faced with a rising epidemic of obesity, perpetuating a preexisting epidemic of diabetes. 

An optimal, comprehensive approach to managing patients with type 2 DM should encompass both the control of dysglycemia and its associated comorbidities, obesity being the key player.⁸ However, clinical  practice is often misaligned with the evidence. For instance, many of our first-line oral treatments for type 2 DM (except for metformin) are associated with weight gain.⁹ With time, control of glycemia becomes more and more ineffective, at which point therapy is intensified with insulin, further exacerbating the weight gain.¹⁰

Therefore, it seems counterintuitive to treat a disease for which obesity is one of the main risk factors with medications that promote weight gain. Yet healthcare providers are faced with a therapeutic dilemma: should they focus their efforts on improving patients’ glycemic control, or should they invest in helping these patients lose weight? Although an ideal approach would incorporate both aspects, the reality is that it is far from practical.

A few issues impinge on integrating weight loss in the care of type 2 DM. Although the American Medical Association recognized obesity as a disease in 2013,¹¹ some providers still perceive obesity as a self-inflicted condition that is due to bad lifestyle and behavior.¹¹ Many clinicians may also have low expectations for patients’ success, and often lack the time and knowledge to intervene regarding nutrition, physical activity, and psychological issues pertinent to the management of obesity in type 2 DM. Therefore, in many cases, it seems less complicated and more rewarding for both patients and physicians to concentrate on improving the HbA1c value rather than investing efforts in weight loss. For diabetic patients with obesity, this could mean that clinicians may prescribe glucose-lowering therapies, such as insulin and sulfonylureas, at the expense of weight gain. Additionally, clinicians often experience the need to provide recommendations more aligned with metrics that dictate reimbursement (eg, HbA1c targets) within healthcare systems that still raise concerns regarding obesity visit reimbursements.

Lastly, the lack of trustworthy or pertinent evidence (lack of comparative effectiveness research) for antiobesity medications may limit their use in daily practice. Physicians have had little confidence in the efficacy of antiobesity drugs, and often raise significant safety concerns, especially after witnessing important fiascos in this field, eg, dexfenfluramine, rimonabant, and sibutramine.^2,12,13

As a result, many of our patients with obesity and type 2 DM may not consider the need for weight loss, and may not even be aware that type 2 DM is caused by obesity and physical inactivity in the first place. Others have accumulated a significant degree of frustration, and have “thrown in the towel” already after unsuccessful weight-loss efforts, many of which were not medically supervised.

For all of the above reasons, both clinicians and patients often concentrate their efforts on treating blood glucose numbers rather than the “obesity-diabetes” as a whole.¹⁴ And as a result, our practices are slowly filling up with patients with obesity and type 2 DM who are treated primarily with insulin, resulting in a progressive (and untreated) obesity and diabetes epidemic.

DRUGS FOR TREATING OBESITY AND TYPE 2 DM

Orlistat

Orlistat (Xenical) is the only weight-loss drug approved by the US Food and Drug Administration (FDA) that acts outside the brain. It inhibits pancreatic lipases, resulting in up to 30% less fat absorption in the gut. Orlistat has been approved for long-term use by the FDA.

Benefits. In the XENical in the Prevention of Diabetes in Obese Subjects study, treatment with orlistat resulted in a significant reduction in the cumulative incidence of type 2 DM after 4 years of treatment (9.0% with placebo vs 6.2% with orlistat), corresponding to a risk reduction of 37.3%.¹⁶ Mean weight loss after 4 years was significantly greater in the orlistat group (5.8 vs 3.0 kg with placebo; P < .001).¹⁶ Other benefits of orlistat included a reduction in low-density lipoprotein cholesterol independent of that expected from change in body weight.¹⁶

Adverse effects include flatulence with discharge and fecal urgency after high-fat dietary indiscretions. Serum levels of fat-soluble vitamins (A, D, E, and K) were lower with orlistat than with placebo,¹⁶ and a fat-soluble vitamin supplement should be taken 2 hours before or after taking orlistat. Serious but very uncommon adverse events such as kidney damage have been reported.¹⁷ Kidney and liver function should be monitored while taking orlistat.

Phentermine (Adipex-P, Lomaira), a sympathomimetic amine, is the most commonly prescribed antiobesity drug in the United States. A schedule IV controlled substance, it is FDA-approved for short-term use (up to 12 weeks). Its primary mechanism of action is mediated by reduction in hunger perception. It was first developed in the 1970s and is available in doses ranging from 8 mg to 37.5 mg daily.¹⁸

Benefits. In a randomized trial, at 28 weeks, weight loss was 1.5 kg with placebo and 5.3 kg with phentermine.¹⁹ No long-term (> 1 year) randomized controlled trials of the effectiveness of phentermine monotherapy in weight loss have been conducted.

Adverse effects. Dizziness, dry mouth, insomnia, constipation, and increase in heart rate were most common.¹⁹

Phentermine is contraindicated in patients with coronary artery disease, congestive heart failure, stroke, and uncontrolled hypertension. Currently, no data exist on the long-term cardiovascular effects of phentermine. We believe phentermine, used in patients at low to intermediate cardiovascular risk, is a useful “jumpstart” tool, in combination with lifestyle changes, to achieve weight loss and improve metabolic values for those with type 2 DM and obesity.

Phentermine is a controlled substance per Ohio law. Patients must be seen once a month by the prescribing provider and prescriptions are limited to a 30-day supply, which must be filled within 7 days of the date of the prescription. Phentermine can only be prescribed for a maximum of 3 months and must be discontinued for 6 months before patients are eligible for a new prescription.

Phentermine and topiramate extended-release

Obesity is a product of complex interactions between several neurohormonal pathways. Approaches simultaneously targeting more than one regulatory pathway have become popular and quite efficient strategies in treating patients with obesity.²⁰ Stemming from such approaches, antiobesity drug combinations such as phentermine and topiramate extended-release (Qsymia) have become increasingly recognized and used in clinical practice. The combination of these 2 medications has been approved for long-term use by the FDA.

Phentermine and topiramate extended-release is a fixed-dose combination that was approved for weight loss in 2012. Topiramate, an anticonvulsant, and phentermine exert their anorexigenic effects through regulating various brain neurotransmitters and result in more weight loss when used together than when either is used alone. Several clinical trials evaluated the efficacy of low doses of this combination in weight loss.

Benefits. In a randomized trial in patients with obesity and cardiometabolic diseases, at 56 weeks, the mean weight loss was:

• 1.2% in the placebo group
• 7.8% in the group receiving phentermine 7.5 mg and topiramate 46 mg
• 9.8% in the group receiving phentermine 15 mg and topiramate 92 mg.²¹

Patients in the active treatment groups also had significant improvements in cardiovascular and metabolic risk factors such as waist circumference, systolic blood pressure, and total cholesterol/high-density lipoprotein cholesterol ratio. At 56 weeks, patients with diabetes and prediabetes taking this preparation had greater reductions in HbA1c values, and fewer prediabetes patients progressed to type 2 DM.²¹

Adverse effects most commonly seen were dry mouth, paresthesia, and constipation.²¹

This combination is contraindicated in pregnancy, patients with recent stroke, uncontrolled hypertension, coronary artery disease, glaucoma, hyperthyroidism, or in patients taking monoamine oxidase inhibitors. Women of childbearing age should be tested for pregnancy before starting therapy, and monthly thereafter, and also be advised to use effective methods of contraception while taking the medication. Topiramate has been associated with the development of renal stones and thus should be used with caution in patients with a history of kidney stones.

Bupropion and naltrexone sustained-release

Bupropion and naltrexone sustained-release (Contrave) is another FDA-approved combination drug for chronic weight management. Bupropion is a dopamine and norepinephrine reuptake inhibitor approved for depression and smoking cessation, and naltrexone is an opioid receptor antagonist approved for treating alcohol and opioid dependence. The combination of these 2 medications has been approved for long-term use by the FDA.

Benefits. In a randomized trial in patients with obesity and type 2 DM, weight loss at 56 weeks was:

• 1.8% with placebo
• 5.0% with naltrexone 32 mg and bupropion 360 mg daily.

Absolute reductions in HbA1c were:

• 0.1% with placebo
• 0.6% with naltrexone-bupropion.

Improvements were also seen in other cardiometabolic risk factors such as triglyceride and high-density lipoprotein cholesterol levels.²²

Adverse effects. The most common adverse effect leading to drug discontinuation was nausea. Other adverse effects reported were constipation, headache, vomiting, and dizziness.²²

Naltrexone-bupropion is contraindicated in patients with a history of seizure disorder or a diagnosis of anorexia nervosa or bulimia, or who are on chronic opioid therapy.

Diethylpropion

Diethylpropion (Tenuate, Tenuate Dospan) is a central nervous system stimulant similar to bupropion in its structure. It was approved by the FDA for treating obesity in 1959. It should be used as part of a short-term weight-loss plan, along with a low-calorie diet. Diethylpropion is also a controlled substance and, as with phentermine therapy, patients are required to be seen once a month by their prescriber. Diethylpropion cannot be prescribed for more than 3 months.

Benefits. Weight loss in a randomized trial at 6 months:

• 3.2% with placebo
• 9.8% with diethylpropion 50 mg twice a day.²³

After 6 months, all participants received diethylpropion in an open-label extension for an additional 6 months. At 12 months, the mean weight loss produced by diethylpropion was 10.6%.²³ No differences in heart rate, blood pressure, electrocardiographic results, or psychiatric evaluations were observed.

Adverse effects. As with phentermine, common side effects of diethylpropion include insomnia, dry mouth, dizziness, headache, mild increases in blood pressure, and palpitations.²³

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA for chronic weight management in June 2012. It exerts its effects through binding selectively to central 5-HT_2C serotonin receptors, with poor affinity for 5-HT_2A and 5-HT_2B receptors. Nonselective serotoninergic agents, including fenfluramine and dexfenfluramine, were withdrawn from the market in 1997 after being reported to be associated with valvular heart abnormalities.²⁴ Lorcaserin has been approved for long-term use by the FDA.

Benefits. Mean weight loss at 1 year in the Behavioral Modification and Lorcaserin for Overweight and Obesity Management in Diabetes Mellitus trial²⁵ was:

• 1.5% with placebo
• 5.0% with lorcaserin 10 mg once daily
• 4.5% with lorcaserin 10 mg twice daily.

Absolute reductions in HbA1c values were:

• 0.4% with placebo
• 0.9% with lorcaserin 10 mg once daily
• 1.0% with lorcaserin 10 mg twice daily.

Absolute reductions in fasting plasma glucose values were:

• 11.9 mg/dL with placebo
• 27.4 mg/dL with lorcaserin 10 mg once daily
• 28.4 mg/dL with lorcaserin 10 mg twice daily.²⁵

Adverse effects. The most common adverse effects were headache, dizziness, and fatigue. There was no significant increase in valvulopathy on echocardiography of participants receiving lorcaserin compared with placebo.²⁵

Liraglutide (Saxenda, Victoza) is a glucagon-like peptide-1 (GLP-1) receptor agonist. Native GLP-1 is a hormone secreted by intestinal L cells in response to consumption of fat and carbohydrate-rich foods. It stimulates the release of insulin and suppresses any inappropriately elevated postprandial glucagon levels. In addition to its effect on glucose metabolism, GLP-1 also reduces appetite and delays gastric emptying in humans.²⁶ Unlike the extremely short half-life of native GLP-1 (estimated at 1 to 2 minutes), liraglutide has a half-life of 13 hours, allowing it to be given once daily.²⁶ Liraglutide medication has been approved for long-term use by the FDA.

Benefits. The Liraglutide Effect and Action in Diabetes 1–5 studies compared the effects of liraglutide monotherapy with antidiabetic oral medications or insulin, as well as in combination with antidiabetic oral agents. Liraglutide (Victoza) at doses approved for type 2 DM of 1.2 mg and 1.8 mg daily had significant effects in reducing HbA1c by 0.48% to 1.84% and weight by 2.5 kg to 4 kg.^27,28 At a dose of 3.0 mg, liraglutide (Saxenda) is approved for chronic weight management. This dose of liraglutide has been shown to be effective and safe in patients with type 2 DM and obesity.

In the 56-week SCALE Diabetes trial,²⁹ liraglutide at a dose of 3.0 mg resulted in 6.0% weight reduction, compared with 2.0% in the placebo group. Of participants receiving 3.0 mg of liraglutide, 54.3% achieved more than 5% weight loss at 56 weeks compared with 21.4% with placebo. Liraglutide also resulted in significant improvements in HbA1c (mean change −1.3% vs −0.3% with placebo), fasting and postprandial glucose levels, and fasting glucagon levels.²⁹

The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results trial has shown liraglutide to significantly reduce rates of major cardiovascular events (first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in patients with elevated cardiovascular risk factors.³⁰ These findings make liraglutide a favorable choice for high-risk patients with type 2 DM, obesity, and cardiovascular disease.

It is important to indicate that if a 5% weight loss is not achieved by 3 months with any of these weight-loss medications, it would be reasonable to stop the medication and consider switching to a different medication. These medications work best when combined with diet and increased physical activity. Weight-loss medications should never be used during pregnancy.

Women of childbearing age should be advised to use effective contraception methods while taking any of the above antiobesity medications.

Diabetes medications associated with weight loss: Metformin and SGLT-2 inhibitors

Although not FDA-approved for weight management, metformin has anorexigenic effects that aid in weight loss. It also inhibits hepatic glucose production and improves peripheral insulin sensitivity, making it a useful agent in patients with type 2 DM and obesity.

A meta-analysis of 31 trials showed that metformin reduced body mass index by 5.3% compared with placebo.³¹ Metformin should be considered as a first-line agent in obese patients with type 2 DM.

In healthy people, nearly all glucose is filtered in the glomerulus, but then 98% of it is reabsorbed in the proximal tubule by sodium-glucose cotransporter-2 (SGLT-2). Drugs that inhibit SGLT-2 increase urinary glucose excretion and, as a result, help control hyperglycemia. Another, off-label effect of excreting more glucose is weight loss: a sustained weight loss of about 3 kg to 5 kg in clinical studies.³² Although they can be used as monotherapies, SGLT-2 inhibitors are usually used as add-on therapies in patients with type 2 DM.

AN ALGORITHM FOR TREATMENT

First, we believe that lifestyle interventions by optimization of nutrition and physical activity should be the cornerstone therapy in the management plan of any patient with type 2 DM and obesity. These interventions are best implemented through a comprehensive, multidisciplinary approach that integrates the care of dietitians, physical therapists, exercise physiologists, psychologists, and social workers.³³ Patients need also to be seen frequently, ie, at least once every 3 months. The possibility of seeing patients in group-shared medical appointments on a monthly basis could also be considered. 

We also believe that metformin should be added early in the course of treatment for its known benefits of improving insulin sensitivity and suppressing appetite. Target HbA1c goals and body weight in patients with type 2 diabetes and obesity should be tailored to the individual based on age, general health status, risk of hypoglycemia, capacity to do physical activity, and associated comorbidities. If no improvements are seen (HbA1c > 7% and < 3 % weight loss) despite lifestyle changes and the addition of metformin, the possibility of adding a GLP-1 receptor agonist or an SGLT-2 inhibitor as a second-line therapy should be considered. Both classes of medications aid in lowering HbA1c and promote further weight loss.

If no clinical progress is achieved at 3 months, the possibility of adding an FDA-approved weight-loss medication, as discussed above, should be strongly considered. Of note, this algorithm targets different endogenous pathways for weight loss and thus minimizes weight regain through compensatory mechanisms.

Patient-centered care has become a core quality measure in our healthcare systems and a key to our patients’ success. The decision to start an antiobesity drug should therefore reflect careful consideration of medical and personal patient issues, all of which are valued differently by patients.³⁴

Individualized therapy is even more relevant among patients suffering from a significant burden of disease. About 80% of patients with diabetes live with at least 1 other medical condition,³⁵ and each of these patients spends over 2 hours a day, on average, following doctors’ recommendations.³⁶ If antiobesity medications are prescribed without careful consideration of the patient’s preexisting workload, they will be destined to fail. Therefore, it becomes crucial to first account for the patient’s ability to cope with therapy intensification. This requires careful deliberation between healthcare providers and patients, in aims of targeting a weight-loss plan that fits patients’ goals and is aligned with providers’ expectations.

Healthcare systems also play a key role in supporting better conversations about obesity in type 2 DM patients. They could implement multifaceted initiatives to promote shared decision-making and the use of decision aids to advance patient-centered obesity practices.³⁷ Policymakers could redesign quality measures aimed at capturing the quality of obesity conversations, and develop policies that support better education for clinicians regarding the importance of addressing obesity with adequate communication and patient-centered skills. Guidelines are often too disease-specific and do not consider comorbidities in their context when providing recommendations.³⁸ Thus, diabetes societies should respond to the need to guide care for patients with diabetes and its comorbidities, particularly obesity.

CONCLUSIONS

Obesity is a serious global health issue and a leading risk factor for type 2 DM. Lifestyle measures are the cornerstone of preventing and treating obesity and type 2 DM. Emerging data support the effectiveness of intensive, interdisciplinary weight-loss programs in patients with diabetes. The use of antiobesity drugs should be considered in patients who have not achieved adequate responses to lifestyle interventions. Medications should be tailored to the individual’s health risks and metabolic and psychobehavioral characteristics. In many cases, the addition of weight-loss drugs will help accomplish and maintain the recommended 10% weight reduction, resulting in improvement in glycemic control and significant reduction in cardiovascular risk factors. New studies combining antiobesity and antidiabetes medications in the context of lifestyle interventions will help define the optimal therapeutic approach for patients with type 2 DM and obesity.

Obesity is a leading public health concern, affecting nearly 60 million adult Americans.¹ It is a major risk factor for the development of insulin resistance and type 2 diabetes mellitus (DM).² More than 90% of patients with type 2 DM have obesity, and obesity is a major obstacle to achieving long-term glycemic control.³

Clinical studies have demonstrated that a 6- to 7-kg increase in body weight increases the risk of developing type 2 DM by 50%, while a 5-kg loss reduces the risk by a similar amount.⁴ As a result, most patients who have a body mass index greater than 40 kg/m² suffer from type 2 DM.⁵ Strong evidence exists that bariatric surgery and its resulting weight loss has positive effects on fasting blood sugar, hemoglobin A1c (HbA1c), lipid profiles, and other metabolic variables.⁶

When combined, obesity and type 2 DM carry a significant burden of micro- and macrovascular complications such as retinopathy, nephropathy, neuropathy, and cardiovascular disease. As a result, a high prevalence of morbidity and mortality is seen among patients with obesity and type 2 DM; those between the ages of 51 and 61 have a 7-times higher mortality rate compared with nonobese normoglycemic people, and patients with diabetes alone have a 2.6-times higher mortality rate.⁷

A DILEMMA IN THE CLINIC: FOCUS ON THE SUGAR OR THE WEIGHT?

Although type 2 DM and obesity go hand in hand, clinicians tend to focus on the sugar and neglect the weight, concentrating their efforts on improving blood glucose indices, and prescribing in many instances medications that cause weight gain. As a result, we are faced with a rising epidemic of obesity, perpetuating a preexisting epidemic of diabetes. 

An optimal, comprehensive approach to managing patients with type 2 DM should encompass both the control of dysglycemia and its associated comorbidities, obesity being the key player.⁸ However, clinical  practice is often misaligned with the evidence. For instance, many of our first-line oral treatments for type 2 DM (except for metformin) are associated with weight gain.⁹ With time, control of glycemia becomes more and more ineffective, at which point therapy is intensified with insulin, further exacerbating the weight gain.¹⁰

Therefore, it seems counterintuitive to treat a disease for which obesity is one of the main risk factors with medications that promote weight gain. Yet healthcare providers are faced with a therapeutic dilemma: should they focus their efforts on improving patients’ glycemic control, or should they invest in helping these patients lose weight? Although an ideal approach would incorporate both aspects, the reality is that it is far from practical.

A few issues impinge on integrating weight loss in the care of type 2 DM. Although the American Medical Association recognized obesity as a disease in 2013,¹¹ some providers still perceive obesity as a self-inflicted condition that is due to bad lifestyle and behavior.¹¹ Many clinicians may also have low expectations for patients’ success, and often lack the time and knowledge to intervene regarding nutrition, physical activity, and psychological issues pertinent to the management of obesity in type 2 DM. Therefore, in many cases, it seems less complicated and more rewarding for both patients and physicians to concentrate on improving the HbA1c value rather than investing efforts in weight loss. For diabetic patients with obesity, this could mean that clinicians may prescribe glucose-lowering therapies, such as insulin and sulfonylureas, at the expense of weight gain. Additionally, clinicians often experience the need to provide recommendations more aligned with metrics that dictate reimbursement (eg, HbA1c targets) within healthcare systems that still raise concerns regarding obesity visit reimbursements.

Lastly, the lack of trustworthy or pertinent evidence (lack of comparative effectiveness research) for antiobesity medications may limit their use in daily practice. Physicians have had little confidence in the efficacy of antiobesity drugs, and often raise significant safety concerns, especially after witnessing important fiascos in this field, eg, dexfenfluramine, rimonabant, and sibutramine.^2,12,13

As a result, many of our patients with obesity and type 2 DM may not consider the need for weight loss, and may not even be aware that type 2 DM is caused by obesity and physical inactivity in the first place. Others have accumulated a significant degree of frustration, and have “thrown in the towel” already after unsuccessful weight-loss efforts, many of which were not medically supervised.

For all of the above reasons, both clinicians and patients often concentrate their efforts on treating blood glucose numbers rather than the “obesity-diabetes” as a whole.¹⁴ And as a result, our practices are slowly filling up with patients with obesity and type 2 DM who are treated primarily with insulin, resulting in a progressive (and untreated) obesity and diabetes epidemic.

DRUGS FOR TREATING OBESITY AND TYPE 2 DM

Orlistat

Orlistat (Xenical) is the only weight-loss drug approved by the US Food and Drug Administration (FDA) that acts outside the brain. It inhibits pancreatic lipases, resulting in up to 30% less fat absorption in the gut. Orlistat has been approved for long-term use by the FDA.

Benefits. In the XENical in the Prevention of Diabetes in Obese Subjects study, treatment with orlistat resulted in a significant reduction in the cumulative incidence of type 2 DM after 4 years of treatment (9.0% with placebo vs 6.2% with orlistat), corresponding to a risk reduction of 37.3%.¹⁶ Mean weight loss after 4 years was significantly greater in the orlistat group (5.8 vs 3.0 kg with placebo; P < .001).¹⁶ Other benefits of orlistat included a reduction in low-density lipoprotein cholesterol independent of that expected from change in body weight.¹⁶

Adverse effects include flatulence with discharge and fecal urgency after high-fat dietary indiscretions. Serum levels of fat-soluble vitamins (A, D, E, and K) were lower with orlistat than with placebo,¹⁶ and a fat-soluble vitamin supplement should be taken 2 hours before or after taking orlistat. Serious but very uncommon adverse events such as kidney damage have been reported.¹⁷ Kidney and liver function should be monitored while taking orlistat.

Phentermine (Adipex-P, Lomaira), a sympathomimetic amine, is the most commonly prescribed antiobesity drug in the United States. A schedule IV controlled substance, it is FDA-approved for short-term use (up to 12 weeks). Its primary mechanism of action is mediated by reduction in hunger perception. It was first developed in the 1970s and is available in doses ranging from 8 mg to 37.5 mg daily.¹⁸

Benefits. In a randomized trial, at 28 weeks, weight loss was 1.5 kg with placebo and 5.3 kg with phentermine.¹⁹ No long-term (> 1 year) randomized controlled trials of the effectiveness of phentermine monotherapy in weight loss have been conducted.

Adverse effects. Dizziness, dry mouth, insomnia, constipation, and increase in heart rate were most common.¹⁹

Phentermine is contraindicated in patients with coronary artery disease, congestive heart failure, stroke, and uncontrolled hypertension. Currently, no data exist on the long-term cardiovascular effects of phentermine. We believe phentermine, used in patients at low to intermediate cardiovascular risk, is a useful “jumpstart” tool, in combination with lifestyle changes, to achieve weight loss and improve metabolic values for those with type 2 DM and obesity.

Phentermine is a controlled substance per Ohio law. Patients must be seen once a month by the prescribing provider and prescriptions are limited to a 30-day supply, which must be filled within 7 days of the date of the prescription. Phentermine can only be prescribed for a maximum of 3 months and must be discontinued for 6 months before patients are eligible for a new prescription.

Phentermine and topiramate extended-release

Obesity is a product of complex interactions between several neurohormonal pathways. Approaches simultaneously targeting more than one regulatory pathway have become popular and quite efficient strategies in treating patients with obesity.²⁰ Stemming from such approaches, antiobesity drug combinations such as phentermine and topiramate extended-release (Qsymia) have become increasingly recognized and used in clinical practice. The combination of these 2 medications has been approved for long-term use by the FDA.

Phentermine and topiramate extended-release is a fixed-dose combination that was approved for weight loss in 2012. Topiramate, an anticonvulsant, and phentermine exert their anorexigenic effects through regulating various brain neurotransmitters and result in more weight loss when used together than when either is used alone. Several clinical trials evaluated the efficacy of low doses of this combination in weight loss.

Benefits. In a randomized trial in patients with obesity and cardiometabolic diseases, at 56 weeks, the mean weight loss was:

• 1.2% in the placebo group
• 7.8% in the group receiving phentermine 7.5 mg and topiramate 46 mg
• 9.8% in the group receiving phentermine 15 mg and topiramate 92 mg.²¹

Patients in the active treatment groups also had significant improvements in cardiovascular and metabolic risk factors such as waist circumference, systolic blood pressure, and total cholesterol/high-density lipoprotein cholesterol ratio. At 56 weeks, patients with diabetes and prediabetes taking this preparation had greater reductions in HbA1c values, and fewer prediabetes patients progressed to type 2 DM.²¹

Adverse effects most commonly seen were dry mouth, paresthesia, and constipation.²¹

This combination is contraindicated in pregnancy, patients with recent stroke, uncontrolled hypertension, coronary artery disease, glaucoma, hyperthyroidism, or in patients taking monoamine oxidase inhibitors. Women of childbearing age should be tested for pregnancy before starting therapy, and monthly thereafter, and also be advised to use effective methods of contraception while taking the medication. Topiramate has been associated with the development of renal stones and thus should be used with caution in patients with a history of kidney stones.

Bupropion and naltrexone sustained-release

Bupropion and naltrexone sustained-release (Contrave) is another FDA-approved combination drug for chronic weight management. Bupropion is a dopamine and norepinephrine reuptake inhibitor approved for depression and smoking cessation, and naltrexone is an opioid receptor antagonist approved for treating alcohol and opioid dependence. The combination of these 2 medications has been approved for long-term use by the FDA.

Benefits. In a randomized trial in patients with obesity and type 2 DM, weight loss at 56 weeks was:

• 1.8% with placebo
• 5.0% with naltrexone 32 mg and bupropion 360 mg daily.

Absolute reductions in HbA1c were:

• 0.1% with placebo
• 0.6% with naltrexone-bupropion.

Improvements were also seen in other cardiometabolic risk factors such as triglyceride and high-density lipoprotein cholesterol levels.²²

Adverse effects. The most common adverse effect leading to drug discontinuation was nausea. Other adverse effects reported were constipation, headache, vomiting, and dizziness.²²

Naltrexone-bupropion is contraindicated in patients with a history of seizure disorder or a diagnosis of anorexia nervosa or bulimia, or who are on chronic opioid therapy.

Diethylpropion

Diethylpropion (Tenuate, Tenuate Dospan) is a central nervous system stimulant similar to bupropion in its structure. It was approved by the FDA for treating obesity in 1959. It should be used as part of a short-term weight-loss plan, along with a low-calorie diet. Diethylpropion is also a controlled substance and, as with phentermine therapy, patients are required to be seen once a month by their prescriber. Diethylpropion cannot be prescribed for more than 3 months.

Benefits. Weight loss in a randomized trial at 6 months:

• 3.2% with placebo
• 9.8% with diethylpropion 50 mg twice a day.²³

After 6 months, all participants received diethylpropion in an open-label extension for an additional 6 months. At 12 months, the mean weight loss produced by diethylpropion was 10.6%.²³ No differences in heart rate, blood pressure, electrocardiographic results, or psychiatric evaluations were observed.

Adverse effects. As with phentermine, common side effects of diethylpropion include insomnia, dry mouth, dizziness, headache, mild increases in blood pressure, and palpitations.²³

Lorcaserin

Lorcaserin (Belviq) was approved by the FDA for chronic weight management in June 2012. It exerts its effects through binding selectively to central 5-HT_2C serotonin receptors, with poor affinity for 5-HT_2A and 5-HT_2B receptors. Nonselective serotoninergic agents, including fenfluramine and dexfenfluramine, were withdrawn from the market in 1997 after being reported to be associated with valvular heart abnormalities.²⁴ Lorcaserin has been approved for long-term use by the FDA.

Benefits. Mean weight loss at 1 year in the Behavioral Modification and Lorcaserin for Overweight and Obesity Management in Diabetes Mellitus trial²⁵ was:

• 1.5% with placebo
• 5.0% with lorcaserin 10 mg once daily
• 4.5% with lorcaserin 10 mg twice daily.

Absolute reductions in HbA1c values were:

• 0.4% with placebo
• 0.9% with lorcaserin 10 mg once daily
• 1.0% with lorcaserin 10 mg twice daily.

Absolute reductions in fasting plasma glucose values were:

• 11.9 mg/dL with placebo
• 27.4 mg/dL with lorcaserin 10 mg once daily
• 28.4 mg/dL with lorcaserin 10 mg twice daily.²⁵

Adverse effects. The most common adverse effects were headache, dizziness, and fatigue. There was no significant increase in valvulopathy on echocardiography of participants receiving lorcaserin compared with placebo.²⁵

Liraglutide (Saxenda, Victoza) is a glucagon-like peptide-1 (GLP-1) receptor agonist. Native GLP-1 is a hormone secreted by intestinal L cells in response to consumption of fat and carbohydrate-rich foods. It stimulates the release of insulin and suppresses any inappropriately elevated postprandial glucagon levels. In addition to its effect on glucose metabolism, GLP-1 also reduces appetite and delays gastric emptying in humans.²⁶ Unlike the extremely short half-life of native GLP-1 (estimated at 1 to 2 minutes), liraglutide has a half-life of 13 hours, allowing it to be given once daily.²⁶ Liraglutide medication has been approved for long-term use by the FDA.

Benefits. The Liraglutide Effect and Action in Diabetes 1–5 studies compared the effects of liraglutide monotherapy with antidiabetic oral medications or insulin, as well as in combination with antidiabetic oral agents. Liraglutide (Victoza) at doses approved for type 2 DM of 1.2 mg and 1.8 mg daily had significant effects in reducing HbA1c by 0.48% to 1.84% and weight by 2.5 kg to 4 kg.^27,28 At a dose of 3.0 mg, liraglutide (Saxenda) is approved for chronic weight management. This dose of liraglutide has been shown to be effective and safe in patients with type 2 DM and obesity.

In the 56-week SCALE Diabetes trial,²⁹ liraglutide at a dose of 3.0 mg resulted in 6.0% weight reduction, compared with 2.0% in the placebo group. Of participants receiving 3.0 mg of liraglutide, 54.3% achieved more than 5% weight loss at 56 weeks compared with 21.4% with placebo. Liraglutide also resulted in significant improvements in HbA1c (mean change −1.3% vs −0.3% with placebo), fasting and postprandial glucose levels, and fasting glucagon levels.²⁹

The Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results trial has shown liraglutide to significantly reduce rates of major cardiovascular events (first occurrence of cardiovascular death, nonfatal myocardial infarction, or nonfatal stroke) in patients with elevated cardiovascular risk factors.³⁰ These findings make liraglutide a favorable choice for high-risk patients with type 2 DM, obesity, and cardiovascular disease.

It is important to indicate that if a 5% weight loss is not achieved by 3 months with any of these weight-loss medications, it would be reasonable to stop the medication and consider switching to a different medication. These medications work best when combined with diet and increased physical activity. Weight-loss medications should never be used during pregnancy.

Women of childbearing age should be advised to use effective contraception methods while taking any of the above antiobesity medications.

Diabetes medications associated with weight loss: Metformin and SGLT-2 inhibitors

Although not FDA-approved for weight management, metformin has anorexigenic effects that aid in weight loss. It also inhibits hepatic glucose production and improves peripheral insulin sensitivity, making it a useful agent in patients with type 2 DM and obesity.

A meta-analysis of 31 trials showed that metformin reduced body mass index by 5.3% compared with placebo.³¹ Metformin should be considered as a first-line agent in obese patients with type 2 DM.

In healthy people, nearly all glucose is filtered in the glomerulus, but then 98% of it is reabsorbed in the proximal tubule by sodium-glucose cotransporter-2 (SGLT-2). Drugs that inhibit SGLT-2 increase urinary glucose excretion and, as a result, help control hyperglycemia. Another, off-label effect of excreting more glucose is weight loss: a sustained weight loss of about 3 kg to 5 kg in clinical studies.³² Although they can be used as monotherapies, SGLT-2 inhibitors are usually used as add-on therapies in patients with type 2 DM.

AN ALGORITHM FOR TREATMENT

First, we believe that lifestyle interventions by optimization of nutrition and physical activity should be the cornerstone therapy in the management plan of any patient with type 2 DM and obesity. These interventions are best implemented through a comprehensive, multidisciplinary approach that integrates the care of dietitians, physical therapists, exercise physiologists, psychologists, and social workers.³³ Patients need also to be seen frequently, ie, at least once every 3 months. The possibility of seeing patients in group-shared medical appointments on a monthly basis could also be considered. 

We also believe that metformin should be added early in the course of treatment for its known benefits of improving insulin sensitivity and suppressing appetite. Target HbA1c goals and body weight in patients with type 2 diabetes and obesity should be tailored to the individual based on age, general health status, risk of hypoglycemia, capacity to do physical activity, and associated comorbidities. If no improvements are seen (HbA1c > 7% and < 3 % weight loss) despite lifestyle changes and the addition of metformin, the possibility of adding a GLP-1 receptor agonist or an SGLT-2 inhibitor as a second-line therapy should be considered. Both classes of medications aid in lowering HbA1c and promote further weight loss.

If no clinical progress is achieved at 3 months, the possibility of adding an FDA-approved weight-loss medication, as discussed above, should be strongly considered. Of note, this algorithm targets different endogenous pathways for weight loss and thus minimizes weight regain through compensatory mechanisms.

Patient-centered care has become a core quality measure in our healthcare systems and a key to our patients’ success. The decision to start an antiobesity drug should therefore reflect careful consideration of medical and personal patient issues, all of which are valued differently by patients.³⁴

Individualized therapy is even more relevant among patients suffering from a significant burden of disease. About 80% of patients with diabetes live with at least 1 other medical condition,³⁵ and each of these patients spends over 2 hours a day, on average, following doctors’ recommendations.³⁶ If antiobesity medications are prescribed without careful consideration of the patient’s preexisting workload, they will be destined to fail. Therefore, it becomes crucial to first account for the patient’s ability to cope with therapy intensification. This requires careful deliberation between healthcare providers and patients, in aims of targeting a weight-loss plan that fits patients’ goals and is aligned with providers’ expectations.

Healthcare systems also play a key role in supporting better conversations about obesity in type 2 DM patients. They could implement multifaceted initiatives to promote shared decision-making and the use of decision aids to advance patient-centered obesity practices.³⁷ Policymakers could redesign quality measures aimed at capturing the quality of obesity conversations, and develop policies that support better education for clinicians regarding the importance of addressing obesity with adequate communication and patient-centered skills. Guidelines are often too disease-specific and do not consider comorbidities in their context when providing recommendations.³⁸ Thus, diabetes societies should respond to the need to guide care for patients with diabetes and its comorbidities, particularly obesity.

CONCLUSIONS

Obesity is a serious global health issue and a leading risk factor for type 2 DM. Lifestyle measures are the cornerstone of preventing and treating obesity and type 2 DM. Emerging data support the effectiveness of intensive, interdisciplinary weight-loss programs in patients with diabetes. The use of antiobesity drugs should be considered in patients who have not achieved adequate responses to lifestyle interventions. Medications should be tailored to the individual’s health risks and metabolic and psychobehavioral characteristics. In many cases, the addition of weight-loss drugs will help accomplish and maintain the recommended 10% weight reduction, resulting in improvement in glycemic control and significant reduction in cardiovascular risk factors. New studies combining antiobesity and antidiabetes medications in the context of lifestyle interventions will help define the optimal therapeutic approach for patients with type 2 DM and obesity.

Data from Bays et al.1

LIMITATIONS OF LIFESTYLE MANAGEMENT AND MEDICATIONS

First-line therapy with lifestyle management and second-line therapy with medications, including oral agents and insulin, are the mainstays of type 2 DM therapy. Although these approaches have reduced hyperglycemia and cardiovascular mortality, many patients have poor glycemic control and develop severe diabetes-related complications. A study using data from the National Health and Nutrition Examination Survey (N = 4,926) to evaluate success rates of lifestyle management plus drug therapy found that just 53% of patients with type 2 DM maintained a hemoglobin A1c (HbA1c) below 7%.⁶ Similarly, only 51% of those patients achieved a systolic and diastolic blood pressure less than 130/80 mm Hg, and only 56% achieved a low-density lipoprotein cholesterol level less than 100 mg/dL. Altogether, only 19% of the study cohort achieved all 3 therapy targets. Documented limitations of lifestyle counseling and drug therapy include behavior maladaptation, limitations in drug potency, nonadherence to medications, adverse effects, and economic deterrents.⁷

METABOLIC SURGERY FOR TYPE 2 DM

For patients with obesity and type 2 DM in whom lifestyle management and medications do not achieve desired treatment goals, bariatric surgery has emerged as the most effective treatment for attaining significant and durable weight loss. These gastrointestinal (GI) procedures, which reduce gastric volume with or without rerouting nutrient flow through the small intestine, were developed to yield long-term weight loss in patients with severe obesity. It is now known that they also cause dramatic improvement or remission of obesity-related comorbidities, especially type 2 DM. Research has shown that these effects are not only secondary to weight loss but also depend on neuroendocrine mechanisms secondary to changes in GI physiology. For these reasons, bariatric surgery is increasingly used with the primary intent to treat type 2 DM or metabolic disease, a practice referred to as metabolic surgery.

For more than 2 decades, indications for metabolic surgery reflected guidelines from a 1991 National Institutes of Health (NIH) consensus conference, which suggested considering surgery only in patients with a BMI of 40 kg/m² or greater or a BMI of 35 kg/m² or greater and significant obesity-related comorbidities.¹¹ Guidelines published in 2013 expanded the recommendations to include adults with a BMI of at least 35 kg/m² and an obesity-related comorbidity, such as diabetes, who are motivated to lose weight.⁴ These recommendations were primarily designed to guide the use of surgery as a weight-loss intervention for severe obesity. However, guidelines published in 2016 support use of metabolic surgery as a specific treatment for type 2 DM.⁵

Potential mechanisms resolving type 2 DM: More than weight loss

Bariatric surgery has been shown to have profound glucoregulatory effects. These include rapid improvement in hyperglycemia and reduction in exogenous insulin requirements that occur early after surgery and before the patient has any significant weight loss.^12,13 Additionally, experiments in rodents showed that changes to GI anatomy can directly influence glucose homeostasis, independently of weight loss and caloric restriction.¹⁴

Although the exact molecular mechanisms underlying the effects of metabolic surgery on diabetes are not fully understood, many factors appear to play a role, including changes in bile acid metabolism, GI tract nutrient sensing, glucose utilization, insulin resistance, and intestinal microbiomes.¹⁵ These changes, acting through peripheral or central pathways, or perhaps both, lead to reduced hepatic glucose production, increased tissue glucose uptake, improved insulin sensitivity, and enhanced beta-cell function. A constellation of gut-derived neuro­endocrine changes, rather than a single overarching mechanism, is the likely mediator of postoperative glycemic improvement, with the contributing factors varying according to the surgical procedure.

Weight loss

Long-term reduction of excess body fat is a major goal of metabolic and bariatric surgery. Weight loss is usually expressed as either the percent of weight loss or the percent of excess weight loss (ie, weight loss above ideal weight). A meta-analysis of mostly short-term weight-loss outcomes (ie, < 5 years) from more than 22,000 procedures found an overall mean excess weight loss of 47.5% for patients who underwent LAGB, 61.6% for RYGB, 68.2% for vertical-banded gastroplasty, and 70.1% for BPD-DS.¹⁶ Vertical-banded gastroplasty differs from LAGB in that both a band and staples are used to create a small stomach pouch. Excess weight loss for SG generally averages 50% to 55%, which is intermediate between LAGB and RYGB.^17,18

The Swedish Obese Subjects study (N = 4,047), a prospective study of bariatric surgery vs nonsurgical weight management of severely obese patients (BMI > 34), is the largest weight-loss study with the longest follow-up.¹⁹ At 20 years, the mean weight loss was 26% for gastric bypass, 18% for vertical-banded gastroplasty, 13% for gastric banding, and 1% for controls. A 10-year study in 1,787 severely obese patients (BMI ≥ 35) who underwent RYGB had 21% more weight loss from their baseline weight than the nonsurgical match.²⁰ At 4-year follow-up in 2,410 patients, there were significant variations in weight loss depending on the procedure: 27.5% for RYGB, 17.8% for SG, and 10.6% in LAGB. Between 2% and 31% regained weight back to baseline: 30.5% for LAGB, 14.6% for SG, and 2.5% for RYGB.²⁰ In contrast, long-term medical (nonsurgical) weight loss rarely exceeds 5%, even with intensive lifestyle intervention.²¹

Diabetes remission, cardiovascular risk factors, glycemic control

A meta-analysis of 19 mostly observational studies (N = 4,070 patients) reported an overall type 2 DM remission rate of 78% after bariatric surgery with 1 to 3 years of follow-up.²² Resolution or remission was typically defined as becoming “nondiabetic” with normal HbA1c without medications. In the Swedish Obese Subjects study, the remission rate was 72% at 2 years and 36% at 10 years compared with 21% and 13%, respectively, for the nonsurgical controls (P < .001).²³ Bariatric surgery was also markedly more effective than nonsurgical treatment in preventing type 2 DM, with a relative risk reduction of 78%.

A systematic review published in 2012 evaluated long-term cardiovascular risk reduction after bariatric surgery in 73 studies and 19,543 patients.²⁴ At a mean follow-up of 57.8 months, the average excess weight loss for all procedures was 54% and rates of remission or improvement were 63% for hypertension, 73% for type 2 DM, and 65% for hyperlipidemia. Results from 12 cohort-matched, nonrandomized studies comparing bariatric surgery vs nonsurgical controls suggest that improvements in surrogate disease markers such as HbA1c, blood pressure, lipids, and body weight after surgery translate to reduced macrovascular and microvascular events and death.²⁵ One of these studies involving male veterans who were mostly at high cardiovascular risk reported a 42% reduction in mortality at 10 years compared with medical therapy.²⁶

In the Swedish Obese Subjects study, the mortality rate from cardiovascular disease in the bariatric surgical group was lower than for control patients (adjusted hazard ratio, 0.47; P = .002) despite a greater prevalence of smoking and higher baseline weights and blood pressures in the surgical cohort.¹⁹ For patients with type 2 DM in this study, surgery was associated with a 50% reduction in microvascular complications.²⁷ After 15 years of follow-up, the cumulative incidence of microvascular complications was 41.8 per 1,000 person-years for control patients and 20.6 per 1,000 person-years in the surgery group (hazard ratio, 0.44; P < .001).

These observational, nonrandomized study data suggest that in patients with type 2 DM, bariatric surgery is significantly better than medical management alone in improving glycemic control, reducing cardiovascular risk factors, and lowering long-term morbidity and mortality associated with type 2 DM.

METABOLIC SURGERY: CLINICAL TRIALS

Collectively, these RCTs showed that surgery was significantly superior to medical treatment in reaching the designated glycemic target (P < .05 for all). The one exception showed that diabetes remission for LAGB vs medical treatment was 33% and 23%, respectively.⁴¹ This result might be due to patients in this study having advanced type 2 DM (HbA1c 8.2% ± 1.2%, with 40% on insulin), and they likely had reduced beta-cell function. Overall, surgery decreased HbA1c by 2% to 3.5%, whereas medical treatment lowered it by only 1% to 1.5%. Most of these studies also showed superiority of surgery over medical treatment in achieving secondary end points such as weight loss, remission of metabolic syndrome, reduction in diabetes and cardiovascular medications, and improvement in triglycerides, lipids, and quality of life. Results were mixed in terms of improvements in systolic and diastolic blood pressure or low-density lipoproteins after surgery vs medical treatment, but many studies did show a corresponding reduction in medication usage.

Durability of the effects of surgery was demonstrated in a 5-year study that showed superior and durable weight loss and glycemic control (remission) with both RYGB and BPD in severely obese patients (BMI ≥ 35) vs medical therapy.³² Similarly, Schauer et al⁴³ showed that RYGB and SG were more effective than intensive medical therapy in improving or, in some cases, resolving hyperglycemia for 5 years. In the RCTs, patients who preoperatively had shorter duration of diabetes, lower HbA1c levels, no insulin requirement, and more postoperative weight loss were more likely to achieve diabetes remission.

Although previous guidelines and payer coverage policies had limited metabolic surgery to severely obese patients (BMI ≥ 35 kg/m²), nearly all RCTs showed that the surgical procedures, especially RYGB and SG, were equally effective in patients with BMI 30 to 35 kg/m². This is particularly important given that most patients with type 2 DM have a BMI less than 35 kg/m². The effect of surgery in these patients with mild obesity is also durable out to at least 5 years.⁴³

No RCT was sufficiently powered to detect differences in macrovascular or microvascular complications or death, especially at the relatively short follow-up, and no such differences have been detected thus far. The STAMPEDE (Surgical Therapy and Medications Potentially Eradicate Diabetes Efficiently) trial⁴³ showed that bariatric surgery (RYGB or SG) did not appear to worsen or improve retinopathy outcomes at 5 years compared with intensive medical management.

Surgical complications

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Nutritional deficiencies

Postoperative nutritional deficiencies are typically associated with diminished nutrient intake or the malabsorptive effect of bariatric procedures. They are more common after RYGB and BPD-DS and less common after SG and LAGB. In addition, there is a high prevalence of nutritional deficiencies (35%–80%) in patients seeking bariatric surgery; thus, poor preoperative nutrition may be a factor in the development of postoperative deficiencies. Common preoperative nutrient deficiencies are vitamin A (11%), vitamin B₁₂ (13%), vitamin D (40%), zinc (30%), iron (16%), ferritin (9%), selenium (58%), and folate (6%).⁵¹ Recommendations are to assess for these deficiencies and correct any identified before surgery.

Mild anemia after bariatric procedures is common, occurring in 15% to 20% of cases, and it is believed to result from reduced absorption of iron and B₁₂, as well from pre-existing iron deficiency anemia in premenopausal patients.⁵² Deficiencies in trace minerals (selenium, zinc, and copper) and vitamins (B₁₂, B₁, A, E, D, and K) can occur after bariatric procedures, especially after BPD-DS.⁵³ Nutrient deficiencies can be prevented or corrected with appropriate vitamin, iron, and calcium supplementation.⁵⁴ 

Bone mineral density may decrease after bariatric surgery (14% in the proximal femur).⁵⁵ Reduced mechanical loading after weight loss, reduced consumption and malabsorption of micronutrients (calcium, vitamin D), and neurohormonal alterations are potential underlying mechanisms of bone mineral density reduction after bariatric surgery. Rates of bone fracture and osteoporosis are not well delineated, raising questions about whether bone loss after bariatric surgery is clinically relevant or a functional adaptation to skeletal unloading. However, the extreme malabsorptive procedures of BPD-DS have been associated with severe calcium and vitamin D deficiencies, leading to decreased bone mineral density and osteoporosis.

Protein malnutrition also can occur after these extreme malabsorptive procedures. Patients require postoperative oral protein supplementation (80–100 g/day) and lifelong monitoring for nutritional complications after these procedures.⁵⁶

Additional complications

Other late complications of bariatric surgery that are less clear in incidence and cause include kidney stones, alcohol abuse, depression, and suicide. One study of patients after RYGB (N = 4,690) reported a significantly higher prevalence of kidney stones than in obese controls: 7.5% vs 4.6%, respectively.⁵⁷ Proposed causes of kidney stone formation following bariatric surgery include hyperoxaluria, hypocitraturia, and elevated urine acidity.⁵⁸

The prevalence of alcohol-use disorder after bariatric surgery ranges from 7.6% to 11.8% and appears to be higher in patients with a history of alcohol use.⁵⁹ Paradoxically, while bariatric surgery has been shown to significantly decrease depression,⁶⁰ some studies suggest that a slight increase in the risk of suicide may occur,⁶¹ while others do not.⁶² A recent review concluded that accurate rates of suicide after bariatric surgery are not known, but practitioners should be aware of this concern and appropriately screen and counsel their patients.⁶³

Although the 12 RCTs reported in Table 1 were not powered to detect differences in treatment-related complications, the overall rates of complications were consistent with those in observational studies.⁹ The most common surgical complications were anemia (15%), need for reoperation (8%), and GI (5%–10%). The 30-day surgical mortality rate was 0.2% (1 death) among the 465 surgical patients. Complications were not limited to the surgical patients. In the medical-treatment control group of the STAMPEDE trial,³⁰ anemia (16%) and weight gain (16%) were common. Investigators reported challenges with medication compliance, including adverse effects leading to discontinuation of medications. Mild hypoglycemia was common, with no significant differences between the surgical and medical treatment groups.

METABOLIC SURGERY: COST EFFECTIVENESS

The cost of bariatric procedures varies considerably but, in general, ranges from $20,000 to $30,000, similar to the cost of cholecystectomy, hysterectomy, and colectomy. Retrospective analyses and modeling studies indicate that metabolic surgery is cost-effective and may present a cost savings in patients with type 2 DM, with a break-even time between 5 and 10 years.^64,65 The cost savings, largely based on assumptions of long-term effectiveness and safety, result from reductions in medication use, outpatient care costs, and long-term complications of type 2 DM.

Until recently, there was no clear national or international consensus on the role of metabolic surgery in treating type 2 DM. In 2015, the 2nd Diabetes Surgery Summit (DSS-II) Consensus Conference published guidelines that were endorsed by more than 50 diabetes and medical organizations.⁵ The recommendations cover many clinically relevant issues, including patient selection, preoperative evaluation, choice of procedure, and postoperative follow-up. The consensus conference delegates concluded that there is sufficient evidence demonstrating that metabolic surgery achieves excellent glycemic control and reduces cardiovascular risk factors.

The treatment algorithm from DSS-II incorporates appropriate use of all 3 treatment modalities: lifestyle intervention, drug therapy, and surgery (Figure 5).⁵ The 2017 Standards of Care for Diabetes from the American Diabetes Association include those key indications in the recommendations for metabolic surgery (Table 3).²

SUMMARY

The safety of metabolic surgery has significantly improved with the advent of laparoscopic surgery and recent national quality improvement initiatives that have made gastric bypass and SG as safe as cholecystectomy and appendectomy. Although observational studies suggest that metabolic surgery is associated with a reduction in cardiovascular and diabetes complications and mortality, these observations have not been confirmed in long-term RCTs.

Based on the published evidence, metabolic surgery is now endorsed as a standard treatment option, which provides patients and practitioners with a powerful tool to help combat the life-impairing effects of type 2 DM.

Data from Bays et al.1

LIMITATIONS OF LIFESTYLE MANAGEMENT AND MEDICATIONS

First-line therapy with lifestyle management and second-line therapy with medications, including oral agents and insulin, are the mainstays of type 2 DM therapy. Although these approaches have reduced hyperglycemia and cardiovascular mortality, many patients have poor glycemic control and develop severe diabetes-related complications. A study using data from the National Health and Nutrition Examination Survey (N = 4,926) to evaluate success rates of lifestyle management plus drug therapy found that just 53% of patients with type 2 DM maintained a hemoglobin A1c (HbA1c) below 7%.⁶ Similarly, only 51% of those patients achieved a systolic and diastolic blood pressure less than 130/80 mm Hg, and only 56% achieved a low-density lipoprotein cholesterol level less than 100 mg/dL. Altogether, only 19% of the study cohort achieved all 3 therapy targets. Documented limitations of lifestyle counseling and drug therapy include behavior maladaptation, limitations in drug potency, nonadherence to medications, adverse effects, and economic deterrents.⁷

METABOLIC SURGERY FOR TYPE 2 DM

For patients with obesity and type 2 DM in whom lifestyle management and medications do not achieve desired treatment goals, bariatric surgery has emerged as the most effective treatment for attaining significant and durable weight loss. These gastrointestinal (GI) procedures, which reduce gastric volume with or without rerouting nutrient flow through the small intestine, were developed to yield long-term weight loss in patients with severe obesity. It is now known that they also cause dramatic improvement or remission of obesity-related comorbidities, especially type 2 DM. Research has shown that these effects are not only secondary to weight loss but also depend on neuroendocrine mechanisms secondary to changes in GI physiology. For these reasons, bariatric surgery is increasingly used with the primary intent to treat type 2 DM or metabolic disease, a practice referred to as metabolic surgery.

For more than 2 decades, indications for metabolic surgery reflected guidelines from a 1991 National Institutes of Health (NIH) consensus conference, which suggested considering surgery only in patients with a BMI of 40 kg/m² or greater or a BMI of 35 kg/m² or greater and significant obesity-related comorbidities.¹¹ Guidelines published in 2013 expanded the recommendations to include adults with a BMI of at least 35 kg/m² and an obesity-related comorbidity, such as diabetes, who are motivated to lose weight.⁴ These recommendations were primarily designed to guide the use of surgery as a weight-loss intervention for severe obesity. However, guidelines published in 2016 support use of metabolic surgery as a specific treatment for type 2 DM.⁵

Potential mechanisms resolving type 2 DM: More than weight loss

Bariatric surgery has been shown to have profound glucoregulatory effects. These include rapid improvement in hyperglycemia and reduction in exogenous insulin requirements that occur early after surgery and before the patient has any significant weight loss.^12,13 Additionally, experiments in rodents showed that changes to GI anatomy can directly influence glucose homeostasis, independently of weight loss and caloric restriction.¹⁴

Although the exact molecular mechanisms underlying the effects of metabolic surgery on diabetes are not fully understood, many factors appear to play a role, including changes in bile acid metabolism, GI tract nutrient sensing, glucose utilization, insulin resistance, and intestinal microbiomes.¹⁵ These changes, acting through peripheral or central pathways, or perhaps both, lead to reduced hepatic glucose production, increased tissue glucose uptake, improved insulin sensitivity, and enhanced beta-cell function. A constellation of gut-derived neuro­endocrine changes, rather than a single overarching mechanism, is the likely mediator of postoperative glycemic improvement, with the contributing factors varying according to the surgical procedure.

Weight loss

Long-term reduction of excess body fat is a major goal of metabolic and bariatric surgery. Weight loss is usually expressed as either the percent of weight loss or the percent of excess weight loss (ie, weight loss above ideal weight). A meta-analysis of mostly short-term weight-loss outcomes (ie, < 5 years) from more than 22,000 procedures found an overall mean excess weight loss of 47.5% for patients who underwent LAGB, 61.6% for RYGB, 68.2% for vertical-banded gastroplasty, and 70.1% for BPD-DS.¹⁶ Vertical-banded gastroplasty differs from LAGB in that both a band and staples are used to create a small stomach pouch. Excess weight loss for SG generally averages 50% to 55%, which is intermediate between LAGB and RYGB.^17,18

The Swedish Obese Subjects study (N = 4,047), a prospective study of bariatric surgery vs nonsurgical weight management of severely obese patients (BMI > 34), is the largest weight-loss study with the longest follow-up.¹⁹ At 20 years, the mean weight loss was 26% for gastric bypass, 18% for vertical-banded gastroplasty, 13% for gastric banding, and 1% for controls. A 10-year study in 1,787 severely obese patients (BMI ≥ 35) who underwent RYGB had 21% more weight loss from their baseline weight than the nonsurgical match.²⁰ At 4-year follow-up in 2,410 patients, there were significant variations in weight loss depending on the procedure: 27.5% for RYGB, 17.8% for SG, and 10.6% in LAGB. Between 2% and 31% regained weight back to baseline: 30.5% for LAGB, 14.6% for SG, and 2.5% for RYGB.²⁰ In contrast, long-term medical (nonsurgical) weight loss rarely exceeds 5%, even with intensive lifestyle intervention.²¹

Diabetes remission, cardiovascular risk factors, glycemic control

A meta-analysis of 19 mostly observational studies (N = 4,070 patients) reported an overall type 2 DM remission rate of 78% after bariatric surgery with 1 to 3 years of follow-up.²² Resolution or remission was typically defined as becoming “nondiabetic” with normal HbA1c without medications. In the Swedish Obese Subjects study, the remission rate was 72% at 2 years and 36% at 10 years compared with 21% and 13%, respectively, for the nonsurgical controls (P < .001).²³ Bariatric surgery was also markedly more effective than nonsurgical treatment in preventing type 2 DM, with a relative risk reduction of 78%.

A systematic review published in 2012 evaluated long-term cardiovascular risk reduction after bariatric surgery in 73 studies and 19,543 patients.²⁴ At a mean follow-up of 57.8 months, the average excess weight loss for all procedures was 54% and rates of remission or improvement were 63% for hypertension, 73% for type 2 DM, and 65% for hyperlipidemia. Results from 12 cohort-matched, nonrandomized studies comparing bariatric surgery vs nonsurgical controls suggest that improvements in surrogate disease markers such as HbA1c, blood pressure, lipids, and body weight after surgery translate to reduced macrovascular and microvascular events and death.²⁵ One of these studies involving male veterans who were mostly at high cardiovascular risk reported a 42% reduction in mortality at 10 years compared with medical therapy.²⁶

In the Swedish Obese Subjects study, the mortality rate from cardiovascular disease in the bariatric surgical group was lower than for control patients (adjusted hazard ratio, 0.47; P = .002) despite a greater prevalence of smoking and higher baseline weights and blood pressures in the surgical cohort.¹⁹ For patients with type 2 DM in this study, surgery was associated with a 50% reduction in microvascular complications.²⁷ After 15 years of follow-up, the cumulative incidence of microvascular complications was 41.8 per 1,000 person-years for control patients and 20.6 per 1,000 person-years in the surgery group (hazard ratio, 0.44; P < .001).

These observational, nonrandomized study data suggest that in patients with type 2 DM, bariatric surgery is significantly better than medical management alone in improving glycemic control, reducing cardiovascular risk factors, and lowering long-term morbidity and mortality associated with type 2 DM.

METABOLIC SURGERY: CLINICAL TRIALS

Collectively, these RCTs showed that surgery was significantly superior to medical treatment in reaching the designated glycemic target (P < .05 for all). The one exception showed that diabetes remission for LAGB vs medical treatment was 33% and 23%, respectively.⁴¹ This result might be due to patients in this study having advanced type 2 DM (HbA1c 8.2% ± 1.2%, with 40% on insulin), and they likely had reduced beta-cell function. Overall, surgery decreased HbA1c by 2% to 3.5%, whereas medical treatment lowered it by only 1% to 1.5%. Most of these studies also showed superiority of surgery over medical treatment in achieving secondary end points such as weight loss, remission of metabolic syndrome, reduction in diabetes and cardiovascular medications, and improvement in triglycerides, lipids, and quality of life. Results were mixed in terms of improvements in systolic and diastolic blood pressure or low-density lipoproteins after surgery vs medical treatment, but many studies did show a corresponding reduction in medication usage.

Durability of the effects of surgery was demonstrated in a 5-year study that showed superior and durable weight loss and glycemic control (remission) with both RYGB and BPD in severely obese patients (BMI ≥ 35) vs medical therapy.³² Similarly, Schauer et al⁴³ showed that RYGB and SG were more effective than intensive medical therapy in improving or, in some cases, resolving hyperglycemia for 5 years. In the RCTs, patients who preoperatively had shorter duration of diabetes, lower HbA1c levels, no insulin requirement, and more postoperative weight loss were more likely to achieve diabetes remission.

Although previous guidelines and payer coverage policies had limited metabolic surgery to severely obese patients (BMI ≥ 35 kg/m²), nearly all RCTs showed that the surgical procedures, especially RYGB and SG, were equally effective in patients with BMI 30 to 35 kg/m². This is particularly important given that most patients with type 2 DM have a BMI less than 35 kg/m². The effect of surgery in these patients with mild obesity is also durable out to at least 5 years.⁴³

No RCT was sufficiently powered to detect differences in macrovascular or microvascular complications or death, especially at the relatively short follow-up, and no such differences have been detected thus far. The STAMPEDE (Surgical Therapy and Medications Potentially Eradicate Diabetes Efficiently) trial⁴³ showed that bariatric surgery (RYGB or SG) did not appear to worsen or improve retinopathy outcomes at 5 years compared with intensive medical management.

Surgical complications

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Nutritional deficiencies

Postoperative nutritional deficiencies are typically associated with diminished nutrient intake or the malabsorptive effect of bariatric procedures. They are more common after RYGB and BPD-DS and less common after SG and LAGB. In addition, there is a high prevalence of nutritional deficiencies (35%–80%) in patients seeking bariatric surgery; thus, poor preoperative nutrition may be a factor in the development of postoperative deficiencies. Common preoperative nutrient deficiencies are vitamin A (11%), vitamin B₁₂ (13%), vitamin D (40%), zinc (30%), iron (16%), ferritin (9%), selenium (58%), and folate (6%).⁵¹ Recommendations are to assess for these deficiencies and correct any identified before surgery.

Mild anemia after bariatric procedures is common, occurring in 15% to 20% of cases, and it is believed to result from reduced absorption of iron and B₁₂, as well from pre-existing iron deficiency anemia in premenopausal patients.⁵² Deficiencies in trace minerals (selenium, zinc, and copper) and vitamins (B₁₂, B₁, A, E, D, and K) can occur after bariatric procedures, especially after BPD-DS.⁵³ Nutrient deficiencies can be prevented or corrected with appropriate vitamin, iron, and calcium supplementation.⁵⁴ 

Bone mineral density may decrease after bariatric surgery (14% in the proximal femur).⁵⁵ Reduced mechanical loading after weight loss, reduced consumption and malabsorption of micronutrients (calcium, vitamin D), and neurohormonal alterations are potential underlying mechanisms of bone mineral density reduction after bariatric surgery. Rates of bone fracture and osteoporosis are not well delineated, raising questions about whether bone loss after bariatric surgery is clinically relevant or a functional adaptation to skeletal unloading. However, the extreme malabsorptive procedures of BPD-DS have been associated with severe calcium and vitamin D deficiencies, leading to decreased bone mineral density and osteoporosis.

Protein malnutrition also can occur after these extreme malabsorptive procedures. Patients require postoperative oral protein supplementation (80–100 g/day) and lifelong monitoring for nutritional complications after these procedures.⁵⁶

Additional complications

Other late complications of bariatric surgery that are less clear in incidence and cause include kidney stones, alcohol abuse, depression, and suicide. One study of patients after RYGB (N = 4,690) reported a significantly higher prevalence of kidney stones than in obese controls: 7.5% vs 4.6%, respectively.⁵⁷ Proposed causes of kidney stone formation following bariatric surgery include hyperoxaluria, hypocitraturia, and elevated urine acidity.⁵⁸

The prevalence of alcohol-use disorder after bariatric surgery ranges from 7.6% to 11.8% and appears to be higher in patients with a history of alcohol use.⁵⁹ Paradoxically, while bariatric surgery has been shown to significantly decrease depression,⁶⁰ some studies suggest that a slight increase in the risk of suicide may occur,⁶¹ while others do not.⁶² A recent review concluded that accurate rates of suicide after bariatric surgery are not known, but practitioners should be aware of this concern and appropriately screen and counsel their patients.⁶³

Although the 12 RCTs reported in Table 1 were not powered to detect differences in treatment-related complications, the overall rates of complications were consistent with those in observational studies.⁹ The most common surgical complications were anemia (15%), need for reoperation (8%), and GI (5%–10%). The 30-day surgical mortality rate was 0.2% (1 death) among the 465 surgical patients. Complications were not limited to the surgical patients. In the medical-treatment control group of the STAMPEDE trial,³⁰ anemia (16%) and weight gain (16%) were common. Investigators reported challenges with medication compliance, including adverse effects leading to discontinuation of medications. Mild hypoglycemia was common, with no significant differences between the surgical and medical treatment groups.

METABOLIC SURGERY: COST EFFECTIVENESS

The cost of bariatric procedures varies considerably but, in general, ranges from $20,000 to $30,000, similar to the cost of cholecystectomy, hysterectomy, and colectomy. Retrospective analyses and modeling studies indicate that metabolic surgery is cost-effective and may present a cost savings in patients with type 2 DM, with a break-even time between 5 and 10 years.^64,65 The cost savings, largely based on assumptions of long-term effectiveness and safety, result from reductions in medication use, outpatient care costs, and long-term complications of type 2 DM.

Until recently, there was no clear national or international consensus on the role of metabolic surgery in treating type 2 DM. In 2015, the 2nd Diabetes Surgery Summit (DSS-II) Consensus Conference published guidelines that were endorsed by more than 50 diabetes and medical organizations.⁵ The recommendations cover many clinically relevant issues, including patient selection, preoperative evaluation, choice of procedure, and postoperative follow-up. The consensus conference delegates concluded that there is sufficient evidence demonstrating that metabolic surgery achieves excellent glycemic control and reduces cardiovascular risk factors.

The treatment algorithm from DSS-II incorporates appropriate use of all 3 treatment modalities: lifestyle intervention, drug therapy, and surgery (Figure 5).⁵ The 2017 Standards of Care for Diabetes from the American Diabetes Association include those key indications in the recommendations for metabolic surgery (Table 3).²

SUMMARY

The safety of metabolic surgery has significantly improved with the advent of laparoscopic surgery and recent national quality improvement initiatives that have made gastric bypass and SG as safe as cholecystectomy and appendectomy. Although observational studies suggest that metabolic surgery is associated with a reduction in cardiovascular and diabetes complications and mortality, these observations have not been confirmed in long-term RCTs.

Based on the published evidence, metabolic surgery is now endorsed as a standard treatment option, which provides patients and practitioners with a powerful tool to help combat the life-impairing effects of type 2 DM.

Data from Bays et al.1

LIMITATIONS OF LIFESTYLE MANAGEMENT AND MEDICATIONS

First-line therapy with lifestyle management and second-line therapy with medications, including oral agents and insulin, are the mainstays of type 2 DM therapy. Although these approaches have reduced hyperglycemia and cardiovascular mortality, many patients have poor glycemic control and develop severe diabetes-related complications. A study using data from the National Health and Nutrition Examination Survey (N = 4,926) to evaluate success rates of lifestyle management plus drug therapy found that just 53% of patients with type 2 DM maintained a hemoglobin A1c (HbA1c) below 7%.⁶ Similarly, only 51% of those patients achieved a systolic and diastolic blood pressure less than 130/80 mm Hg, and only 56% achieved a low-density lipoprotein cholesterol level less than 100 mg/dL. Altogether, only 19% of the study cohort achieved all 3 therapy targets. Documented limitations of lifestyle counseling and drug therapy include behavior maladaptation, limitations in drug potency, nonadherence to medications, adverse effects, and economic deterrents.⁷

METABOLIC SURGERY FOR TYPE 2 DM

For patients with obesity and type 2 DM in whom lifestyle management and medications do not achieve desired treatment goals, bariatric surgery has emerged as the most effective treatment for attaining significant and durable weight loss. These gastrointestinal (GI) procedures, which reduce gastric volume with or without rerouting nutrient flow through the small intestine, were developed to yield long-term weight loss in patients with severe obesity. It is now known that they also cause dramatic improvement or remission of obesity-related comorbidities, especially type 2 DM. Research has shown that these effects are not only secondary to weight loss but also depend on neuroendocrine mechanisms secondary to changes in GI physiology. For these reasons, bariatric surgery is increasingly used with the primary intent to treat type 2 DM or metabolic disease, a practice referred to as metabolic surgery.

For more than 2 decades, indications for metabolic surgery reflected guidelines from a 1991 National Institutes of Health (NIH) consensus conference, which suggested considering surgery only in patients with a BMI of 40 kg/m² or greater or a BMI of 35 kg/m² or greater and significant obesity-related comorbidities.¹¹ Guidelines published in 2013 expanded the recommendations to include adults with a BMI of at least 35 kg/m² and an obesity-related comorbidity, such as diabetes, who are motivated to lose weight.⁴ These recommendations were primarily designed to guide the use of surgery as a weight-loss intervention for severe obesity. However, guidelines published in 2016 support use of metabolic surgery as a specific treatment for type 2 DM.⁵

Potential mechanisms resolving type 2 DM: More than weight loss

Bariatric surgery has been shown to have profound glucoregulatory effects. These include rapid improvement in hyperglycemia and reduction in exogenous insulin requirements that occur early after surgery and before the patient has any significant weight loss.^12,13 Additionally, experiments in rodents showed that changes to GI anatomy can directly influence glucose homeostasis, independently of weight loss and caloric restriction.¹⁴

Although the exact molecular mechanisms underlying the effects of metabolic surgery on diabetes are not fully understood, many factors appear to play a role, including changes in bile acid metabolism, GI tract nutrient sensing, glucose utilization, insulin resistance, and intestinal microbiomes.¹⁵ These changes, acting through peripheral or central pathways, or perhaps both, lead to reduced hepatic glucose production, increased tissue glucose uptake, improved insulin sensitivity, and enhanced beta-cell function. A constellation of gut-derived neuro­endocrine changes, rather than a single overarching mechanism, is the likely mediator of postoperative glycemic improvement, with the contributing factors varying according to the surgical procedure.

Weight loss

Long-term reduction of excess body fat is a major goal of metabolic and bariatric surgery. Weight loss is usually expressed as either the percent of weight loss or the percent of excess weight loss (ie, weight loss above ideal weight). A meta-analysis of mostly short-term weight-loss outcomes (ie, < 5 years) from more than 22,000 procedures found an overall mean excess weight loss of 47.5% for patients who underwent LAGB, 61.6% for RYGB, 68.2% for vertical-banded gastroplasty, and 70.1% for BPD-DS.¹⁶ Vertical-banded gastroplasty differs from LAGB in that both a band and staples are used to create a small stomach pouch. Excess weight loss for SG generally averages 50% to 55%, which is intermediate between LAGB and RYGB.^17,18

The Swedish Obese Subjects study (N = 4,047), a prospective study of bariatric surgery vs nonsurgical weight management of severely obese patients (BMI > 34), is the largest weight-loss study with the longest follow-up.¹⁹ At 20 years, the mean weight loss was 26% for gastric bypass, 18% for vertical-banded gastroplasty, 13% for gastric banding, and 1% for controls. A 10-year study in 1,787 severely obese patients (BMI ≥ 35) who underwent RYGB had 21% more weight loss from their baseline weight than the nonsurgical match.²⁰ At 4-year follow-up in 2,410 patients, there were significant variations in weight loss depending on the procedure: 27.5% for RYGB, 17.8% for SG, and 10.6% in LAGB. Between 2% and 31% regained weight back to baseline: 30.5% for LAGB, 14.6% for SG, and 2.5% for RYGB.²⁰ In contrast, long-term medical (nonsurgical) weight loss rarely exceeds 5%, even with intensive lifestyle intervention.²¹

Diabetes remission, cardiovascular risk factors, glycemic control

A meta-analysis of 19 mostly observational studies (N = 4,070 patients) reported an overall type 2 DM remission rate of 78% after bariatric surgery with 1 to 3 years of follow-up.²² Resolution or remission was typically defined as becoming “nondiabetic” with normal HbA1c without medications. In the Swedish Obese Subjects study, the remission rate was 72% at 2 years and 36% at 10 years compared with 21% and 13%, respectively, for the nonsurgical controls (P < .001).²³ Bariatric surgery was also markedly more effective than nonsurgical treatment in preventing type 2 DM, with a relative risk reduction of 78%.

A systematic review published in 2012 evaluated long-term cardiovascular risk reduction after bariatric surgery in 73 studies and 19,543 patients.²⁴ At a mean follow-up of 57.8 months, the average excess weight loss for all procedures was 54% and rates of remission or improvement were 63% for hypertension, 73% for type 2 DM, and 65% for hyperlipidemia. Results from 12 cohort-matched, nonrandomized studies comparing bariatric surgery vs nonsurgical controls suggest that improvements in surrogate disease markers such as HbA1c, blood pressure, lipids, and body weight after surgery translate to reduced macrovascular and microvascular events and death.²⁵ One of these studies involving male veterans who were mostly at high cardiovascular risk reported a 42% reduction in mortality at 10 years compared with medical therapy.²⁶

In the Swedish Obese Subjects study, the mortality rate from cardiovascular disease in the bariatric surgical group was lower than for control patients (adjusted hazard ratio, 0.47; P = .002) despite a greater prevalence of smoking and higher baseline weights and blood pressures in the surgical cohort.¹⁹ For patients with type 2 DM in this study, surgery was associated with a 50% reduction in microvascular complications.²⁷ After 15 years of follow-up, the cumulative incidence of microvascular complications was 41.8 per 1,000 person-years for control patients and 20.6 per 1,000 person-years in the surgery group (hazard ratio, 0.44; P < .001).

These observational, nonrandomized study data suggest that in patients with type 2 DM, bariatric surgery is significantly better than medical management alone in improving glycemic control, reducing cardiovascular risk factors, and lowering long-term morbidity and mortality associated with type 2 DM.

METABOLIC SURGERY: CLINICAL TRIALS

Collectively, these RCTs showed that surgery was significantly superior to medical treatment in reaching the designated glycemic target (P < .05 for all). The one exception showed that diabetes remission for LAGB vs medical treatment was 33% and 23%, respectively.⁴¹ This result might be due to patients in this study having advanced type 2 DM (HbA1c 8.2% ± 1.2%, with 40% on insulin), and they likely had reduced beta-cell function. Overall, surgery decreased HbA1c by 2% to 3.5%, whereas medical treatment lowered it by only 1% to 1.5%. Most of these studies also showed superiority of surgery over medical treatment in achieving secondary end points such as weight loss, remission of metabolic syndrome, reduction in diabetes and cardiovascular medications, and improvement in triglycerides, lipids, and quality of life. Results were mixed in terms of improvements in systolic and diastolic blood pressure or low-density lipoproteins after surgery vs medical treatment, but many studies did show a corresponding reduction in medication usage.

Durability of the effects of surgery was demonstrated in a 5-year study that showed superior and durable weight loss and glycemic control (remission) with both RYGB and BPD in severely obese patients (BMI ≥ 35) vs medical therapy.³² Similarly, Schauer et al⁴³ showed that RYGB and SG were more effective than intensive medical therapy in improving or, in some cases, resolving hyperglycemia for 5 years. In the RCTs, patients who preoperatively had shorter duration of diabetes, lower HbA1c levels, no insulin requirement, and more postoperative weight loss were more likely to achieve diabetes remission.

Although previous guidelines and payer coverage policies had limited metabolic surgery to severely obese patients (BMI ≥ 35 kg/m²), nearly all RCTs showed that the surgical procedures, especially RYGB and SG, were equally effective in patients with BMI 30 to 35 kg/m². This is particularly important given that most patients with type 2 DM have a BMI less than 35 kg/m². The effect of surgery in these patients with mild obesity is also durable out to at least 5 years.⁴³

No RCT was sufficiently powered to detect differences in macrovascular or microvascular complications or death, especially at the relatively short follow-up, and no such differences have been detected thus far. The STAMPEDE (Surgical Therapy and Medications Potentially Eradicate Diabetes Efficiently) trial⁴³ showed that bariatric surgery (RYGB or SG) did not appear to worsen or improve retinopathy outcomes at 5 years compared with intensive medical management.

Surgical complications

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Reprinted with permission from John Wiley &amp; Sons (Aminian A, et al. How safe is metabolic/diabetes surgery? Diabetes Obes Metab 2015; 17:198–201.) ©2014 John Wiley &amp; Sons Ltd.

Nutritional deficiencies

Postoperative nutritional deficiencies are typically associated with diminished nutrient intake or the malabsorptive effect of bariatric procedures. They are more common after RYGB and BPD-DS and less common after SG and LAGB. In addition, there is a high prevalence of nutritional deficiencies (35%–80%) in patients seeking bariatric surgery; thus, poor preoperative nutrition may be a factor in the development of postoperative deficiencies. Common preoperative nutrient deficiencies are vitamin A (11%), vitamin B₁₂ (13%), vitamin D (40%), zinc (30%), iron (16%), ferritin (9%), selenium (58%), and folate (6%).⁵¹ Recommendations are to assess for these deficiencies and correct any identified before surgery.

Mild anemia after bariatric procedures is common, occurring in 15% to 20% of cases, and it is believed to result from reduced absorption of iron and B₁₂, as well from pre-existing iron deficiency anemia in premenopausal patients.⁵² Deficiencies in trace minerals (selenium, zinc, and copper) and vitamins (B₁₂, B₁, A, E, D, and K) can occur after bariatric procedures, especially after BPD-DS.⁵³ Nutrient deficiencies can be prevented or corrected with appropriate vitamin, iron, and calcium supplementation.⁵⁴ 

Bone mineral density may decrease after bariatric surgery (14% in the proximal femur).⁵⁵ Reduced mechanical loading after weight loss, reduced consumption and malabsorption of micronutrients (calcium, vitamin D), and neurohormonal alterations are potential underlying mechanisms of bone mineral density reduction after bariatric surgery. Rates of bone fracture and osteoporosis are not well delineated, raising questions about whether bone loss after bariatric surgery is clinically relevant or a functional adaptation to skeletal unloading. However, the extreme malabsorptive procedures of BPD-DS have been associated with severe calcium and vitamin D deficiencies, leading to decreased bone mineral density and osteoporosis.

Protein malnutrition also can occur after these extreme malabsorptive procedures. Patients require postoperative oral protein supplementation (80–100 g/day) and lifelong monitoring for nutritional complications after these procedures.⁵⁶

Additional complications

Other late complications of bariatric surgery that are less clear in incidence and cause include kidney stones, alcohol abuse, depression, and suicide. One study of patients after RYGB (N = 4,690) reported a significantly higher prevalence of kidney stones than in obese controls: 7.5% vs 4.6%, respectively.⁵⁷ Proposed causes of kidney stone formation following bariatric surgery include hyperoxaluria, hypocitraturia, and elevated urine acidity.⁵⁸

The prevalence of alcohol-use disorder after bariatric surgery ranges from 7.6% to 11.8% and appears to be higher in patients with a history of alcohol use.⁵⁹ Paradoxically, while bariatric surgery has been shown to significantly decrease depression,⁶⁰ some studies suggest that a slight increase in the risk of suicide may occur,⁶¹ while others do not.⁶² A recent review concluded that accurate rates of suicide after bariatric surgery are not known, but practitioners should be aware of this concern and appropriately screen and counsel their patients.⁶³

Although the 12 RCTs reported in Table 1 were not powered to detect differences in treatment-related complications, the overall rates of complications were consistent with those in observational studies.⁹ The most common surgical complications were anemia (15%), need for reoperation (8%), and GI (5%–10%). The 30-day surgical mortality rate was 0.2% (1 death) among the 465 surgical patients. Complications were not limited to the surgical patients. In the medical-treatment control group of the STAMPEDE trial,³⁰ anemia (16%) and weight gain (16%) were common. Investigators reported challenges with medication compliance, including adverse effects leading to discontinuation of medications. Mild hypoglycemia was common, with no significant differences between the surgical and medical treatment groups.

METABOLIC SURGERY: COST EFFECTIVENESS

The cost of bariatric procedures varies considerably but, in general, ranges from $20,000 to $30,000, similar to the cost of cholecystectomy, hysterectomy, and colectomy. Retrospective analyses and modeling studies indicate that metabolic surgery is cost-effective and may present a cost savings in patients with type 2 DM, with a break-even time between 5 and 10 years.^64,65 The cost savings, largely based on assumptions of long-term effectiveness and safety, result from reductions in medication use, outpatient care costs, and long-term complications of type 2 DM.

Until recently, there was no clear national or international consensus on the role of metabolic surgery in treating type 2 DM. In 2015, the 2nd Diabetes Surgery Summit (DSS-II) Consensus Conference published guidelines that were endorsed by more than 50 diabetes and medical organizations.⁵ The recommendations cover many clinically relevant issues, including patient selection, preoperative evaluation, choice of procedure, and postoperative follow-up. The consensus conference delegates concluded that there is sufficient evidence demonstrating that metabolic surgery achieves excellent glycemic control and reduces cardiovascular risk factors.

The treatment algorithm from DSS-II incorporates appropriate use of all 3 treatment modalities: lifestyle intervention, drug therapy, and surgery (Figure 5).⁵ The 2017 Standards of Care for Diabetes from the American Diabetes Association include those key indications in the recommendations for metabolic surgery (Table 3).²

SUMMARY

The safety of metabolic surgery has significantly improved with the advent of laparoscopic surgery and recent national quality improvement initiatives that have made gastric bypass and SG as safe as cholecystectomy and appendectomy. Although observational studies suggest that metabolic surgery is associated with a reduction in cardiovascular and diabetes complications and mortality, these observations have not been confirmed in long-term RCTs.

Based on the published evidence, metabolic surgery is now endorsed as a standard treatment option, which provides patients and practitioners with a powerful tool to help combat the life-impairing effects of type 2 DM.

In the United States, 57.9% of patients with diabetes mellitus (DM) have at least 1 diabetes-related complication and 14.3% of patients with diabetes have 3 or more diabetes-related complications.¹ Achieving glycemic control in patients with DM reduces the development and progression of retinopathy, nephropathy, and neuropathy. Aggressive treatment of dyslipidemia and hypertension decreases macrovascular complications.^2–4 The techniques for monitoring blood glucose and the various treatment options available to manage glycemic control in patients with diabetes are reviewed below.

Measuring Glycemic Control

The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of hemoglobin A1c (HbA1c). Continuous glucose monitoring is also available and may be appropriate for select patients, such as patients with brittle diabetes and those using insulin pumps.

Self-monitoring of blood glucose

For patients with type 1 DM and patients with insulin-dependent type 2 DM, self-monitoring of blood glucose allows patients to adjust insulin dosing to prevent hypoglycemia and hyperglycemia.^2,5–7 The American Diabetes Association (ADA) guidelines recommend that patients with type 1 DM self-monitor their glucose:

• Before eating
• At bedtime
• Before exercise
• If hypoglycemia is suspected
• Until hypoglycemia is corrected
• Postprandially upon occasion
• And before critical tasks (ie, driving).⁸

Patients should be educated about how to use real-time blood glucose values to adjust their food intake and medical therapy.

It is commonly recommended that patients with type 2 DM self-monitor their blood glucose levels, but the evidence to support the effectiveness of this practice is inconclusive. Initial studies showed reductions in HbA1c with self-monitoring; however, the inclusion of beneficial health behaviors such as diet and exercise in the analyses makes it difficult to assess the effectiveness of self-monitor blood glucose alone.^2,9

The ADA recommends that nonpregnant adults maintain blood glucose levels of 80 mg/dL to 130 mg/dL preprandial and less than 180 mg/dL postprandial.⁸ The blood glucose goals for patients with gestational diabetes are 95 mg/dL or less preprandial and either 140 mg/dL or less 1-hour postprandial or 120 mg/dL or less 2-hours postprandial.

HbA1c

HbA1c tests reflect the mean blood glucose values over a 3-month period and can predict patients’ risk of microvascular complications.^10,11 The ADA recommends that patients with stable glycemic control have an HbA1c test at least twice a year. Quarterly HbA1c testing is suggested for patients with a recent change in therapy or for patients not meeting their glycemic goals.⁸

Measurement of HbA1c is influenced by the red blood cell turnover rate; therefore, anemia, transfusions, and hemoglobinopathies can cause inaccurate test values. The ADA recommends that nonpregnant adults maintain HbA1c levels near 7%. For patients with diabetes who become pregnant, the goal is HbA1c levels less than 6.0%.⁸ The ADA also recommends that select patients, especially those with a long life expectancy and little comorbidity, adopt glycemic targets near normal levels (HbA1c < 6.5%), providing the target can be achieved without significant hypoglycemia.⁸

Insulin sensitizers

Biguanides (metformin)

Metformin is the only available biguanide. Metformin should be used as a first-line therapy in patients with type 2 DM whenever possible.¹³ Metformin suppresses hepatic glucose output and primarily affects fasting glycemia; however, reduced postprandial glucose concentrations also occur.

The most common side effects of metformin are diarrhea, nausea, and abdominal discomfort. Metformin has the potential to produce very rare but life-threatening lactic acidosis (< 1 in 100,000). The use of metformin is contraindicated in patients with a glomerular filtration rate less than 30 mL/min, with acidosis, hypoxia, or dehydration.⁸

Metformin usually does not lead to hypoglycemia when used as monotherapy. It can lead to weight loss (3%–5% of body weight), and it has been shown to decrease plasma triglyceride concentrations (10%–20%).^8,14,15

Thiazolidinediones

Thiazolidinediones (TZDs) primarily enhance the insulin sensitivity of muscle and fat tissue and mildly enhance insulin sensitivity of the liver. TZDs lower fasting and postprandial blood glucose levels.

Major side effects of TZDs include weight gain, with an increase in subcutaneous adiposity, and fluid retention. Fluid retention typically manifests as peripheral edema, but heart failure can occur on occasion. These agents should be avoided in patients with functional class III or IV heart failure. The PROactive trial of the TZD pioglitazone found that pioglitazone did not increase cardiovascular risk compared with placebo.¹⁶ TZDs have been associated with an increased risk of fractures, particularly in women. When used as monotherapy, TZDs do not cause hypoglycemia. Pioglitazone lowers triglyceride levels, increases high-density lipoprotein cholesterol, and increases the low-density lipoprotein cholesterol particle size.^8,16–18

Insulin secretagogues

Insulin secretagogues such as sulfonylureas and glinides stimulate secretion of insulin from the pancreas regardless of the ambient glucose concentration.

Sulfonylureas

Sulfonylureas lower fasting and postprandial glucose levels. The main side effects include weight gain (about 2 kg upon initiation) and hypoglycemia. The UK Prospective Diabetes Study (UKPDS) trial showed a decrease in microvascular complications with the use of sulfonylureas.¹⁹ Caution should be used in patients with liver or kidney dysfunction or patients who frequently skip meals. Newer, second-generation sulfonylureas (ie, glipizide and glimepiride) may have less risk of hypoglycemia because their action is somewhat glucose dependent.^8,17,19

Glinides

Glinides, which include repaglinide and nateglenide, have a rapid onset of action and a short duration of action, so they are a good option for patients with erratically timed meals. Glinides have a lower risk of hypoglycemia than sulfonylureas. Caution must be used with glinides in patients with liver dysfunction. Dosing is immediately before meals.^8,17

Alpha-glucosidase inhibitors

Alpha-glucosidase inhibitors such as acarbose, miglitol, and voglibose block the enzyme alpha-glucosidase in the cells of the brush border of the small intestine, which delays absorption of carbohydrates. Alpha-glucosidase inhibitors primarily affect postprandial hyperglycemia without causing hypoglycemia. Abdominal cramps, bloating, flatulence, and diarrhea are the most common side effects. Use of alpha-glucosidase inhibitors should be avoided in patients with severe hepatic or renal impairment. Dosing is prior to carbohydrate-containing meals.^8,20

Incretin-based therapies

Therapies that target the incretin hormones to increase insulin production include glucagon-like peptide-1 (GLP-1) receptor agonists and dipeptidyl peptidase-4 (DPP-4) inhibitors.

GLP-1 agonists

Exenatide, liraglutide, albiglutide, and dulaglutide are synthetic analogs of the GLP-1 hormone. GLP-1 is produced in the small intestine; it stimulates insulin secretion and inhibits glucagon secretion in a glucose-dependent manner. It also delays gastric emptying and suppresses appetite through central pathways. GLP-1 agonists primarily decrease postprandial blood glucose levels; however, a moderate reduction in fasting blood glucose and some weight loss can also occur.

The major side effects are gastrointestinal complaints such as nausea, vomiting, and diarrhea. Hypoglycemia does not occur unless GLP-1 analogues are combined with a sulfonylurea or insulin. There is a slightly increased risk of acute pancreatitis in patients using GLP-1 agonist medications, and patients must be warned to discontinue use of these medications if abdominal pain occurs.

Dosing of GLP-1 agonist medications is either twice daily, daily, or weekly by subcutaneous injection.^8,21

DPP-4 inhibitors

DPP-4 is an enzyme that rapidly degrades GLP-1. Suppression of DPP-4 leads to higher levels of insulin secretion and suppression of glucagon secretion in a glucose-dependent manner.

The DPP-4 inhibitors such as linagliptin, sitagliptin, saxagliptin, and alogliptin are given orally once daily. An increased risk of acute pancreatitis has been reported in some patients. Dose reduction is needed in patients with renal impairment for most of these medications.^8,22

SLGT-2 inhibitors

SGLT-2 inhibitors include canagliflozin, dapagliflozin, and empagliflozin and are the newest group of antidiabetic medications. These medications inhibit glucose reabsorption in proximal tubule of the kidney leading to glycosuria, which lowers the blood glucose concentration, lowers blood pressure, and leads to some weight loss. Empagliflozin was shown to be cardioprotective in some patients.²³

SGLT-2 inhibitors are given once a day in the morning and the primary side effects are polyuria and genital yeast infections. These medications are contraindicated in patients with severe end-stage renal disease and those who are on dialysis.^8,24

Pramlintide (amylinomimetics)

Pramlintide, an amylinomimetic, is a synthetic drug that acts like amylin, a hormone secreted by beta cells that suppresses glucagon secretion, slows gastric emptying, and suppresses appetite through central pathways. Pramlintide acts primarily on postprandial blood glucose levels.

The side effects of pramlintide are gastrointestinal complaints, especially nausea. Currently, pramlintide is approved only as an adjunctive therapy with insulin, and it can be used in patients with type 1 DM or type 2 DM. The dose for type 1 DM is 15 µg before each meal subcutaneously, and for type 2 DM it is generally 60 µg before meals.²⁵

Dopamine-receptor agonist (bromocriptine)

Bromocriptine is a central dopamine-receptor agonist, and when given in rapid-release form within 2 hours of awakening in the morning, it improves glycemic control for patients with type 2 DM. The mechanism of action resulting in improved glycemic control is unknown. Studies have demonstrated the cardiovascular safety of bromocriptine.²⁶

Side effects of bromocriptine include hypotension, somnolence, and nausea. Individuals with psychiatric disorders may experience exacerbation while taking bromocriptine. Bromocriptine is taken with food to diminish nausea.²⁷

Insulin

Insulin and insulin analogues remain the most direct method of reducing hyperglycemia. There is no upper limit in dosing for therapeutic effect, so it can be used to bring any HbA1c down to near-normal levels. Other benefits of insulin include reducing triglyceride levels and increasing high-density lipoprotein cholesterol.

Hypoglycemia is a concern with use of insulin, and studies have shown that episodes for which the patient required assistance due to the hypoglycemia occurred between 1 and 3 times per 100 patient-years.¹³ Weight gain can occur after initiation of insulin therapy, and patients typically gain 2 kg to 4 kg.⁸

All patients with type 1 DM require insulin therapy. There are 2 regimens available: basal-bolus and insulin-pump therapy. Patients with type 2 DM often require insulin, which can be combined with oral hypoglycemic agents. Regimens include basal insulin only, twice-daily premixed insulin, basal-bolus therapy, and insulin-pump therapy.²⁸

Basal-bolus therapy

The basal-bolus regimen combines a long-acting agent for basal-insulin needs that is used once or twice daily and a rapid-acting agent for prandial coverage. Traditionally, 50% of the total daily dose is given as basal insulin (detemir, glargine, degludec) and the remaining dose as prandial insulin divided equally before meals (regular, lispro, glulisine, or aspart).

The meal dose of insulin can be fixed, but it is better to determine the dose based on the carbohydrate content of the meal. To do so, patients should be educated about carbohydrate counting and the dose of insulin required to cover the carbohydrate content of the meal. Consultation with a diabetes educator is needed for patients to effectively dose insulin based on the carbohydrate content of meals. Patients are also provided with a sliding scale of supplemental insulin to use as a third component of therapy when the blood glucose level is higher than desired.

The starting total daily insulin dose is typically 0.3 U/kg for patients with type 1 DM and 0.5 U/kg for patients with type 2 DM if no other medications are used. The ADA recommends adding basal insulin at 0.1 to 0.2 U/kg for patients with type 2 DM once they need it. The key to good glycemic control is self-monitoring of blood glucose by the patient and frequent adjustment of the regimen until control is achieved.⁸

Insulin-pump therapy

The insulin pump allows the use of different basal insulin rates at different periods of the day for greater flexibility with daily dosing. The insulin pump also allows administration of the meal bolus as a single discrete bolus or as an extended bolus (square bolus) over a certain period of time, which allows a better match between insulin delivery and glucose absorption from the meal in patients with abnormalities of gastric emptying. Use of an insulin pump should be considered in the following patients:

• Patients unable to achieve target goals with basal-bolus regimens
• Patients with frequent hypoglycemia, dawn phenomenon, or brittle diabetes
• Pregnant patients
• Patients with insulin sensitivity or those requiring more intense monitoring due to complications.

Recently, continuous glucose monitors have been developed that measure interstitial glucose levels. Continuous glucose monitoring has been shown to lower HbA1c in adult patients with type 1 DM.²⁹

Gestational diabetes

In patients with gestational diabetes, insulin therapy is indicated when exercise and nutritional therapy are ineffective in controlling prandial and fasting blood glucose levels. Basal therapy alone may be sufficient, but a basal-bolus regimen is often required.⁸

Summary

• Glycemic control reduces the development and progression of complications of diabetes such as retinopathy, nephropathy, and neuropathy.
• The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of HbA1c.
• Available treatment options to control blood glucose include insulin sensitizers, insulin secretagogues, alpha-glucosidase inhibitors, incretin-based therapies, SGLT-2 inhibitors, amylinomimetics (pramlintide), dopamine-receptor agonist (bromocriptine), and insulin.

In the United States, 57.9% of patients with diabetes mellitus (DM) have at least 1 diabetes-related complication and 14.3% of patients with diabetes have 3 or more diabetes-related complications.¹ Achieving glycemic control in patients with DM reduces the development and progression of retinopathy, nephropathy, and neuropathy. Aggressive treatment of dyslipidemia and hypertension decreases macrovascular complications.^2–4 The techniques for monitoring blood glucose and the various treatment options available to manage glycemic control in patients with diabetes are reviewed below.

Measuring Glycemic Control

The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of hemoglobin A1c (HbA1c). Continuous glucose monitoring is also available and may be appropriate for select patients, such as patients with brittle diabetes and those using insulin pumps.

Self-monitoring of blood glucose

For patients with type 1 DM and patients with insulin-dependent type 2 DM, self-monitoring of blood glucose allows patients to adjust insulin dosing to prevent hypoglycemia and hyperglycemia.^2,5–7 The American Diabetes Association (ADA) guidelines recommend that patients with type 1 DM self-monitor their glucose:

• Before eating
• At bedtime
• Before exercise
• If hypoglycemia is suspected
• Until hypoglycemia is corrected
• Postprandially upon occasion
• And before critical tasks (ie, driving).⁸

Patients should be educated about how to use real-time blood glucose values to adjust their food intake and medical therapy.

It is commonly recommended that patients with type 2 DM self-monitor their blood glucose levels, but the evidence to support the effectiveness of this practice is inconclusive. Initial studies showed reductions in HbA1c with self-monitoring; however, the inclusion of beneficial health behaviors such as diet and exercise in the analyses makes it difficult to assess the effectiveness of self-monitor blood glucose alone.^2,9

The ADA recommends that nonpregnant adults maintain blood glucose levels of 80 mg/dL to 130 mg/dL preprandial and less than 180 mg/dL postprandial.⁸ The blood glucose goals for patients with gestational diabetes are 95 mg/dL or less preprandial and either 140 mg/dL or less 1-hour postprandial or 120 mg/dL or less 2-hours postprandial.

HbA1c

HbA1c tests reflect the mean blood glucose values over a 3-month period and can predict patients’ risk of microvascular complications.^10,11 The ADA recommends that patients with stable glycemic control have an HbA1c test at least twice a year. Quarterly HbA1c testing is suggested for patients with a recent change in therapy or for patients not meeting their glycemic goals.⁸

Measurement of HbA1c is influenced by the red blood cell turnover rate; therefore, anemia, transfusions, and hemoglobinopathies can cause inaccurate test values. The ADA recommends that nonpregnant adults maintain HbA1c levels near 7%. For patients with diabetes who become pregnant, the goal is HbA1c levels less than 6.0%.⁸ The ADA also recommends that select patients, especially those with a long life expectancy and little comorbidity, adopt glycemic targets near normal levels (HbA1c < 6.5%), providing the target can be achieved without significant hypoglycemia.⁸

Insulin sensitizers

Biguanides (metformin)

Metformin is the only available biguanide. Metformin should be used as a first-line therapy in patients with type 2 DM whenever possible.¹³ Metformin suppresses hepatic glucose output and primarily affects fasting glycemia; however, reduced postprandial glucose concentrations also occur.

The most common side effects of metformin are diarrhea, nausea, and abdominal discomfort. Metformin has the potential to produce very rare but life-threatening lactic acidosis (< 1 in 100,000). The use of metformin is contraindicated in patients with a glomerular filtration rate less than 30 mL/min, with acidosis, hypoxia, or dehydration.⁸

Metformin usually does not lead to hypoglycemia when used as monotherapy. It can lead to weight loss (3%–5% of body weight), and it has been shown to decrease plasma triglyceride concentrations (10%–20%).^8,14,15

Thiazolidinediones

Thiazolidinediones (TZDs) primarily enhance the insulin sensitivity of muscle and fat tissue and mildly enhance insulin sensitivity of the liver. TZDs lower fasting and postprandial blood glucose levels.

Major side effects of TZDs include weight gain, with an increase in subcutaneous adiposity, and fluid retention. Fluid retention typically manifests as peripheral edema, but heart failure can occur on occasion. These agents should be avoided in patients with functional class III or IV heart failure. The PROactive trial of the TZD pioglitazone found that pioglitazone did not increase cardiovascular risk compared with placebo.¹⁶ TZDs have been associated with an increased risk of fractures, particularly in women. When used as monotherapy, TZDs do not cause hypoglycemia. Pioglitazone lowers triglyceride levels, increases high-density lipoprotein cholesterol, and increases the low-density lipoprotein cholesterol particle size.^8,16–18

Insulin secretagogues

Insulin secretagogues such as sulfonylureas and glinides stimulate secretion of insulin from the pancreas regardless of the ambient glucose concentration.

Sulfonylureas

Sulfonylureas lower fasting and postprandial glucose levels. The main side effects include weight gain (about 2 kg upon initiation) and hypoglycemia. The UK Prospective Diabetes Study (UKPDS) trial showed a decrease in microvascular complications with the use of sulfonylureas.¹⁹ Caution should be used in patients with liver or kidney dysfunction or patients who frequently skip meals. Newer, second-generation sulfonylureas (ie, glipizide and glimepiride) may have less risk of hypoglycemia because their action is somewhat glucose dependent.^8,17,19

Glinides

Glinides, which include repaglinide and nateglenide, have a rapid onset of action and a short duration of action, so they are a good option for patients with erratically timed meals. Glinides have a lower risk of hypoglycemia than sulfonylureas. Caution must be used with glinides in patients with liver dysfunction. Dosing is immediately before meals.^8,17

Alpha-glucosidase inhibitors

Alpha-glucosidase inhibitors such as acarbose, miglitol, and voglibose block the enzyme alpha-glucosidase in the cells of the brush border of the small intestine, which delays absorption of carbohydrates. Alpha-glucosidase inhibitors primarily affect postprandial hyperglycemia without causing hypoglycemia. Abdominal cramps, bloating, flatulence, and diarrhea are the most common side effects. Use of alpha-glucosidase inhibitors should be avoided in patients with severe hepatic or renal impairment. Dosing is prior to carbohydrate-containing meals.^8,20

Incretin-based therapies

Therapies that target the incretin hormones to increase insulin production include glucagon-like peptide-1 (GLP-1) receptor agonists and dipeptidyl peptidase-4 (DPP-4) inhibitors.

GLP-1 agonists

Exenatide, liraglutide, albiglutide, and dulaglutide are synthetic analogs of the GLP-1 hormone. GLP-1 is produced in the small intestine; it stimulates insulin secretion and inhibits glucagon secretion in a glucose-dependent manner. It also delays gastric emptying and suppresses appetite through central pathways. GLP-1 agonists primarily decrease postprandial blood glucose levels; however, a moderate reduction in fasting blood glucose and some weight loss can also occur.

The major side effects are gastrointestinal complaints such as nausea, vomiting, and diarrhea. Hypoglycemia does not occur unless GLP-1 analogues are combined with a sulfonylurea or insulin. There is a slightly increased risk of acute pancreatitis in patients using GLP-1 agonist medications, and patients must be warned to discontinue use of these medications if abdominal pain occurs.

Dosing of GLP-1 agonist medications is either twice daily, daily, or weekly by subcutaneous injection.^8,21

DPP-4 inhibitors

DPP-4 is an enzyme that rapidly degrades GLP-1. Suppression of DPP-4 leads to higher levels of insulin secretion and suppression of glucagon secretion in a glucose-dependent manner.

The DPP-4 inhibitors such as linagliptin, sitagliptin, saxagliptin, and alogliptin are given orally once daily. An increased risk of acute pancreatitis has been reported in some patients. Dose reduction is needed in patients with renal impairment for most of these medications.^8,22

SLGT-2 inhibitors

SGLT-2 inhibitors include canagliflozin, dapagliflozin, and empagliflozin and are the newest group of antidiabetic medications. These medications inhibit glucose reabsorption in proximal tubule of the kidney leading to glycosuria, which lowers the blood glucose concentration, lowers blood pressure, and leads to some weight loss. Empagliflozin was shown to be cardioprotective in some patients.²³

SGLT-2 inhibitors are given once a day in the morning and the primary side effects are polyuria and genital yeast infections. These medications are contraindicated in patients with severe end-stage renal disease and those who are on dialysis.^8,24

Pramlintide (amylinomimetics)

Pramlintide, an amylinomimetic, is a synthetic drug that acts like amylin, a hormone secreted by beta cells that suppresses glucagon secretion, slows gastric emptying, and suppresses appetite through central pathways. Pramlintide acts primarily on postprandial blood glucose levels.

The side effects of pramlintide are gastrointestinal complaints, especially nausea. Currently, pramlintide is approved only as an adjunctive therapy with insulin, and it can be used in patients with type 1 DM or type 2 DM. The dose for type 1 DM is 15 µg before each meal subcutaneously, and for type 2 DM it is generally 60 µg before meals.²⁵

Dopamine-receptor agonist (bromocriptine)

Bromocriptine is a central dopamine-receptor agonist, and when given in rapid-release form within 2 hours of awakening in the morning, it improves glycemic control for patients with type 2 DM. The mechanism of action resulting in improved glycemic control is unknown. Studies have demonstrated the cardiovascular safety of bromocriptine.²⁶

Side effects of bromocriptine include hypotension, somnolence, and nausea. Individuals with psychiatric disorders may experience exacerbation while taking bromocriptine. Bromocriptine is taken with food to diminish nausea.²⁷

Insulin

Insulin and insulin analogues remain the most direct method of reducing hyperglycemia. There is no upper limit in dosing for therapeutic effect, so it can be used to bring any HbA1c down to near-normal levels. Other benefits of insulin include reducing triglyceride levels and increasing high-density lipoprotein cholesterol.

Hypoglycemia is a concern with use of insulin, and studies have shown that episodes for which the patient required assistance due to the hypoglycemia occurred between 1 and 3 times per 100 patient-years.¹³ Weight gain can occur after initiation of insulin therapy, and patients typically gain 2 kg to 4 kg.⁸

All patients with type 1 DM require insulin therapy. There are 2 regimens available: basal-bolus and insulin-pump therapy. Patients with type 2 DM often require insulin, which can be combined with oral hypoglycemic agents. Regimens include basal insulin only, twice-daily premixed insulin, basal-bolus therapy, and insulin-pump therapy.²⁸

Basal-bolus therapy

The basal-bolus regimen combines a long-acting agent for basal-insulin needs that is used once or twice daily and a rapid-acting agent for prandial coverage. Traditionally, 50% of the total daily dose is given as basal insulin (detemir, glargine, degludec) and the remaining dose as prandial insulin divided equally before meals (regular, lispro, glulisine, or aspart).

The meal dose of insulin can be fixed, but it is better to determine the dose based on the carbohydrate content of the meal. To do so, patients should be educated about carbohydrate counting and the dose of insulin required to cover the carbohydrate content of the meal. Consultation with a diabetes educator is needed for patients to effectively dose insulin based on the carbohydrate content of meals. Patients are also provided with a sliding scale of supplemental insulin to use as a third component of therapy when the blood glucose level is higher than desired.

The starting total daily insulin dose is typically 0.3 U/kg for patients with type 1 DM and 0.5 U/kg for patients with type 2 DM if no other medications are used. The ADA recommends adding basal insulin at 0.1 to 0.2 U/kg for patients with type 2 DM once they need it. The key to good glycemic control is self-monitoring of blood glucose by the patient and frequent adjustment of the regimen until control is achieved.⁸

Insulin-pump therapy

The insulin pump allows the use of different basal insulin rates at different periods of the day for greater flexibility with daily dosing. The insulin pump also allows administration of the meal bolus as a single discrete bolus or as an extended bolus (square bolus) over a certain period of time, which allows a better match between insulin delivery and glucose absorption from the meal in patients with abnormalities of gastric emptying. Use of an insulin pump should be considered in the following patients:

• Patients unable to achieve target goals with basal-bolus regimens
• Patients with frequent hypoglycemia, dawn phenomenon, or brittle diabetes
• Pregnant patients
• Patients with insulin sensitivity or those requiring more intense monitoring due to complications.

Recently, continuous glucose monitors have been developed that measure interstitial glucose levels. Continuous glucose monitoring has been shown to lower HbA1c in adult patients with type 1 DM.²⁹

Gestational diabetes

In patients with gestational diabetes, insulin therapy is indicated when exercise and nutritional therapy are ineffective in controlling prandial and fasting blood glucose levels. Basal therapy alone may be sufficient, but a basal-bolus regimen is often required.⁸

Summary

• Glycemic control reduces the development and progression of complications of diabetes such as retinopathy, nephropathy, and neuropathy.
• The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of HbA1c.
• Available treatment options to control blood glucose include insulin sensitizers, insulin secretagogues, alpha-glucosidase inhibitors, incretin-based therapies, SGLT-2 inhibitors, amylinomimetics (pramlintide), dopamine-receptor agonist (bromocriptine), and insulin.

In the United States, 57.9% of patients with diabetes mellitus (DM) have at least 1 diabetes-related complication and 14.3% of patients with diabetes have 3 or more diabetes-related complications.¹ Achieving glycemic control in patients with DM reduces the development and progression of retinopathy, nephropathy, and neuropathy. Aggressive treatment of dyslipidemia and hypertension decreases macrovascular complications.^2–4 The techniques for monitoring blood glucose and the various treatment options available to manage glycemic control in patients with diabetes are reviewed below.

Measuring Glycemic Control

The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of hemoglobin A1c (HbA1c). Continuous glucose monitoring is also available and may be appropriate for select patients, such as patients with brittle diabetes and those using insulin pumps.

Self-monitoring of blood glucose

For patients with type 1 DM and patients with insulin-dependent type 2 DM, self-monitoring of blood glucose allows patients to adjust insulin dosing to prevent hypoglycemia and hyperglycemia.^2,5–7 The American Diabetes Association (ADA) guidelines recommend that patients with type 1 DM self-monitor their glucose:

• Before eating
• At bedtime
• Before exercise
• If hypoglycemia is suspected
• Until hypoglycemia is corrected
• Postprandially upon occasion
• And before critical tasks (ie, driving).⁸

Patients should be educated about how to use real-time blood glucose values to adjust their food intake and medical therapy.

It is commonly recommended that patients with type 2 DM self-monitor their blood glucose levels, but the evidence to support the effectiveness of this practice is inconclusive. Initial studies showed reductions in HbA1c with self-monitoring; however, the inclusion of beneficial health behaviors such as diet and exercise in the analyses makes it difficult to assess the effectiveness of self-monitor blood glucose alone.^2,9

The ADA recommends that nonpregnant adults maintain blood glucose levels of 80 mg/dL to 130 mg/dL preprandial and less than 180 mg/dL postprandial.⁸ The blood glucose goals for patients with gestational diabetes are 95 mg/dL or less preprandial and either 140 mg/dL or less 1-hour postprandial or 120 mg/dL or less 2-hours postprandial.

HbA1c

HbA1c tests reflect the mean blood glucose values over a 3-month period and can predict patients’ risk of microvascular complications.^10,11 The ADA recommends that patients with stable glycemic control have an HbA1c test at least twice a year. Quarterly HbA1c testing is suggested for patients with a recent change in therapy or for patients not meeting their glycemic goals.⁸

Measurement of HbA1c is influenced by the red blood cell turnover rate; therefore, anemia, transfusions, and hemoglobinopathies can cause inaccurate test values. The ADA recommends that nonpregnant adults maintain HbA1c levels near 7%. For patients with diabetes who become pregnant, the goal is HbA1c levels less than 6.0%.⁸ The ADA also recommends that select patients, especially those with a long life expectancy and little comorbidity, adopt glycemic targets near normal levels (HbA1c < 6.5%), providing the target can be achieved without significant hypoglycemia.⁸

Insulin sensitizers

Biguanides (metformin)

Metformin is the only available biguanide. Metformin should be used as a first-line therapy in patients with type 2 DM whenever possible.¹³ Metformin suppresses hepatic glucose output and primarily affects fasting glycemia; however, reduced postprandial glucose concentrations also occur.

The most common side effects of metformin are diarrhea, nausea, and abdominal discomfort. Metformin has the potential to produce very rare but life-threatening lactic acidosis (< 1 in 100,000). The use of metformin is contraindicated in patients with a glomerular filtration rate less than 30 mL/min, with acidosis, hypoxia, or dehydration.⁸

Metformin usually does not lead to hypoglycemia when used as monotherapy. It can lead to weight loss (3%–5% of body weight), and it has been shown to decrease plasma triglyceride concentrations (10%–20%).^8,14,15

Thiazolidinediones

Thiazolidinediones (TZDs) primarily enhance the insulin sensitivity of muscle and fat tissue and mildly enhance insulin sensitivity of the liver. TZDs lower fasting and postprandial blood glucose levels.

Major side effects of TZDs include weight gain, with an increase in subcutaneous adiposity, and fluid retention. Fluid retention typically manifests as peripheral edema, but heart failure can occur on occasion. These agents should be avoided in patients with functional class III or IV heart failure. The PROactive trial of the TZD pioglitazone found that pioglitazone did not increase cardiovascular risk compared with placebo.¹⁶ TZDs have been associated with an increased risk of fractures, particularly in women. When used as monotherapy, TZDs do not cause hypoglycemia. Pioglitazone lowers triglyceride levels, increases high-density lipoprotein cholesterol, and increases the low-density lipoprotein cholesterol particle size.^8,16–18

Insulin secretagogues

Insulin secretagogues such as sulfonylureas and glinides stimulate secretion of insulin from the pancreas regardless of the ambient glucose concentration.

Sulfonylureas

Sulfonylureas lower fasting and postprandial glucose levels. The main side effects include weight gain (about 2 kg upon initiation) and hypoglycemia. The UK Prospective Diabetes Study (UKPDS) trial showed a decrease in microvascular complications with the use of sulfonylureas.¹⁹ Caution should be used in patients with liver or kidney dysfunction or patients who frequently skip meals. Newer, second-generation sulfonylureas (ie, glipizide and glimepiride) may have less risk of hypoglycemia because their action is somewhat glucose dependent.^8,17,19

Glinides

Glinides, which include repaglinide and nateglenide, have a rapid onset of action and a short duration of action, so they are a good option for patients with erratically timed meals. Glinides have a lower risk of hypoglycemia than sulfonylureas. Caution must be used with glinides in patients with liver dysfunction. Dosing is immediately before meals.^8,17

Alpha-glucosidase inhibitors

Alpha-glucosidase inhibitors such as acarbose, miglitol, and voglibose block the enzyme alpha-glucosidase in the cells of the brush border of the small intestine, which delays absorption of carbohydrates. Alpha-glucosidase inhibitors primarily affect postprandial hyperglycemia without causing hypoglycemia. Abdominal cramps, bloating, flatulence, and diarrhea are the most common side effects. Use of alpha-glucosidase inhibitors should be avoided in patients with severe hepatic or renal impairment. Dosing is prior to carbohydrate-containing meals.^8,20

Incretin-based therapies

Therapies that target the incretin hormones to increase insulin production include glucagon-like peptide-1 (GLP-1) receptor agonists and dipeptidyl peptidase-4 (DPP-4) inhibitors.

GLP-1 agonists

Exenatide, liraglutide, albiglutide, and dulaglutide are synthetic analogs of the GLP-1 hormone. GLP-1 is produced in the small intestine; it stimulates insulin secretion and inhibits glucagon secretion in a glucose-dependent manner. It also delays gastric emptying and suppresses appetite through central pathways. GLP-1 agonists primarily decrease postprandial blood glucose levels; however, a moderate reduction in fasting blood glucose and some weight loss can also occur.

The major side effects are gastrointestinal complaints such as nausea, vomiting, and diarrhea. Hypoglycemia does not occur unless GLP-1 analogues are combined with a sulfonylurea or insulin. There is a slightly increased risk of acute pancreatitis in patients using GLP-1 agonist medications, and patients must be warned to discontinue use of these medications if abdominal pain occurs.

Dosing of GLP-1 agonist medications is either twice daily, daily, or weekly by subcutaneous injection.^8,21

DPP-4 inhibitors

DPP-4 is an enzyme that rapidly degrades GLP-1. Suppression of DPP-4 leads to higher levels of insulin secretion and suppression of glucagon secretion in a glucose-dependent manner.

The DPP-4 inhibitors such as linagliptin, sitagliptin, saxagliptin, and alogliptin are given orally once daily. An increased risk of acute pancreatitis has been reported in some patients. Dose reduction is needed in patients with renal impairment for most of these medications.^8,22

SLGT-2 inhibitors

SGLT-2 inhibitors include canagliflozin, dapagliflozin, and empagliflozin and are the newest group of antidiabetic medications. These medications inhibit glucose reabsorption in proximal tubule of the kidney leading to glycosuria, which lowers the blood glucose concentration, lowers blood pressure, and leads to some weight loss. Empagliflozin was shown to be cardioprotective in some patients.²³

SGLT-2 inhibitors are given once a day in the morning and the primary side effects are polyuria and genital yeast infections. These medications are contraindicated in patients with severe end-stage renal disease and those who are on dialysis.^8,24

Pramlintide (amylinomimetics)

Pramlintide, an amylinomimetic, is a synthetic drug that acts like amylin, a hormone secreted by beta cells that suppresses glucagon secretion, slows gastric emptying, and suppresses appetite through central pathways. Pramlintide acts primarily on postprandial blood glucose levels.

The side effects of pramlintide are gastrointestinal complaints, especially nausea. Currently, pramlintide is approved only as an adjunctive therapy with insulin, and it can be used in patients with type 1 DM or type 2 DM. The dose for type 1 DM is 15 µg before each meal subcutaneously, and for type 2 DM it is generally 60 µg before meals.²⁵

Dopamine-receptor agonist (bromocriptine)

Bromocriptine is a central dopamine-receptor agonist, and when given in rapid-release form within 2 hours of awakening in the morning, it improves glycemic control for patients with type 2 DM. The mechanism of action resulting in improved glycemic control is unknown. Studies have demonstrated the cardiovascular safety of bromocriptine.²⁶

Side effects of bromocriptine include hypotension, somnolence, and nausea. Individuals with psychiatric disorders may experience exacerbation while taking bromocriptine. Bromocriptine is taken with food to diminish nausea.²⁷

Insulin

Insulin and insulin analogues remain the most direct method of reducing hyperglycemia. There is no upper limit in dosing for therapeutic effect, so it can be used to bring any HbA1c down to near-normal levels. Other benefits of insulin include reducing triglyceride levels and increasing high-density lipoprotein cholesterol.

Hypoglycemia is a concern with use of insulin, and studies have shown that episodes for which the patient required assistance due to the hypoglycemia occurred between 1 and 3 times per 100 patient-years.¹³ Weight gain can occur after initiation of insulin therapy, and patients typically gain 2 kg to 4 kg.⁸

All patients with type 1 DM require insulin therapy. There are 2 regimens available: basal-bolus and insulin-pump therapy. Patients with type 2 DM often require insulin, which can be combined with oral hypoglycemic agents. Regimens include basal insulin only, twice-daily premixed insulin, basal-bolus therapy, and insulin-pump therapy.²⁸

Basal-bolus therapy

The basal-bolus regimen combines a long-acting agent for basal-insulin needs that is used once or twice daily and a rapid-acting agent for prandial coverage. Traditionally, 50% of the total daily dose is given as basal insulin (detemir, glargine, degludec) and the remaining dose as prandial insulin divided equally before meals (regular, lispro, glulisine, or aspart).

The meal dose of insulin can be fixed, but it is better to determine the dose based on the carbohydrate content of the meal. To do so, patients should be educated about carbohydrate counting and the dose of insulin required to cover the carbohydrate content of the meal. Consultation with a diabetes educator is needed for patients to effectively dose insulin based on the carbohydrate content of meals. Patients are also provided with a sliding scale of supplemental insulin to use as a third component of therapy when the blood glucose level is higher than desired.

The starting total daily insulin dose is typically 0.3 U/kg for patients with type 1 DM and 0.5 U/kg for patients with type 2 DM if no other medications are used. The ADA recommends adding basal insulin at 0.1 to 0.2 U/kg for patients with type 2 DM once they need it. The key to good glycemic control is self-monitoring of blood glucose by the patient and frequent adjustment of the regimen until control is achieved.⁸

Insulin-pump therapy

The insulin pump allows the use of different basal insulin rates at different periods of the day for greater flexibility with daily dosing. The insulin pump also allows administration of the meal bolus as a single discrete bolus or as an extended bolus (square bolus) over a certain period of time, which allows a better match between insulin delivery and glucose absorption from the meal in patients with abnormalities of gastric emptying. Use of an insulin pump should be considered in the following patients:

• Patients unable to achieve target goals with basal-bolus regimens
• Patients with frequent hypoglycemia, dawn phenomenon, or brittle diabetes
• Pregnant patients
• Patients with insulin sensitivity or those requiring more intense monitoring due to complications.

Recently, continuous glucose monitors have been developed that measure interstitial glucose levels. Continuous glucose monitoring has been shown to lower HbA1c in adult patients with type 1 DM.²⁹

Gestational diabetes

In patients with gestational diabetes, insulin therapy is indicated when exercise and nutritional therapy are ineffective in controlling prandial and fasting blood glucose levels. Basal therapy alone may be sufficient, but a basal-bolus regimen is often required.⁸

Summary

• Glycemic control reduces the development and progression of complications of diabetes such as retinopathy, nephropathy, and neuropathy.
• The primary techniques available to assess the quality of a patient’s glycemic control are self-monitoring of blood glucose and interval measurement of HbA1c.
• Available treatment options to control blood glucose include insulin sensitizers, insulin secretagogues, alpha-glucosidase inhibitors, incretin-based therapies, SGLT-2 inhibitors, amylinomimetics (pramlintide), dopamine-receptor agonist (bromocriptine), and insulin.

A 71-year-old man was referred to the gastroenterology department for evaluation of 9 months of progressive swallowing difficulties associated with epigastric and chest discomfort.

He was a previous smoker (17 pack-years), with a history of coronary artery disease, hypertension, and cervical spinal stenosis requiring decompressive laminectomy with a postoperative course complicated by episodes of aspiration.

DYSPHAGIA: OROPHARYNGEAL OR ESOPHAGEAL

Difficulty swallowing (dysphagia) can be caused by problems in the oropharynx or in the esophagus. Difficulty initiating a swallow can be thought of as oropharyngeal dysphagia, whereas the intermittent sensation of food stuck in the neck or chest is considered esophageal dysphagia.

Focused questioning can help differentiate oropharyngeal symptoms from esophageal symptoms. For example, difficulty clearing secretions or passing the food bolus beyond the mouth or frequent coughing spells while eating is consistent with oropharyngeal dysphagia and suggests a neurologic cause. Our patient, however, presented with a constellation of symptoms more suggestive of esophageal dysphagia.

When eliciting a history of esophageal symptoms, it is crucial to determine the progression of swallowing difficulty, as well as how it directly relates to eating solids or liquids, or both. Difficulty swallowing solid foods that has progressed over time to include liquids would raise concern for an obstruction such as a stricture, ring, or malignancy. On the other hand, abrupt onset of intermittent dysphagia to both solids and liquids would raise concern for a motility disorder of the esophagus. This patient presented with an abrupt onset of intermittent symptoms to both solids and liquids that was associated with substernal chest pain.

Once coronary disease was ruled out by cardiac biomarker testing, electrocardiography, and a pharmacologic stress test, our patient underwent upper endoscopy, which showed a normal esophageal mucosa without masses or obstruction and no evidence of peptic ulcer disease.

WHAT IS THE NEXT STEP?

When upper endoscopy is negative and cardiac causes and gastroesophageal reflux disease have been ruled out, an esophageal motility disorder should be considered.

1. After obstruction has been ruled out with upper endoscopy, which should be the next step in the investigation of esophageal dysphagia?

• A 24-hour pH recording
• Barium esophagography
• Modified barium swallow
• Computed tomography of the chest

Barium esophagography is the optimal fluoroscopic study to evaluate the esophageal phase of the swallow. This study requires the patient to swallow a thick barium solution and a 13-mm barium pill under video analysis. It is useful early in the investigation of esophageal dysphagia because it can potentially reveal areas of esophageal luminal narrowing not detected endoscopically, as well as detail the rate of esophageal emptying.¹

The modified barium swallow, which is performed with the assistance of a speech pathologist, is similar but only shows the oropharynx as far as the cervical esophagus. Therefore, it would be the best fluoroscopic test to assess patients with possible aspiration or oropharyngeal dysphagia, whereas barium esophagography would be the test of choice in evaluating esophageal dysmotility or mechanical obstruction.

pH testing may be helpful in diagnosing gastroesophageal reflux disease but is less helpful in the evaluation of dysphagia.

Computed tomography of the chest may be useful to evaluate for extrinsic compression of the esophagus, but it is not the best next step in the evaluation of dysphagia.

Our patient underwent barium esophagography, which revealed tertiary contractions in the mid and distal esophagus with slight narrowing of the lower cervical esophagus (Figure 1). (Primary contractions are elicited when initiating a swallow that propels the food bolus through the esophagus, while secondary contractions follow in response to esophageal distention to move all remaining esophageal contents from the thoracic esophagus. Tertiary contractions are abnormal, nonpropulsive, spontaneous contractions of the esophageal body that are initiated without swallowing.²)

EOSINOPHILIC ESOPHAGITIS

Histologic study of biopsies of the mid and distal esophagus from our patient’s upper endoscopy revealed 5 eosinophils per high-power field.

2. Does this patient meet the criteria for the diagnosis of eosinophilic esophagitis?

• Yes
• No

No. Having eosinophils in the esophagus is not enough to diagnose eosinophilic esophagitis, as eosinophils are also common in patients with gastroesophageal reflux disease.

Eosinophilic esophagitis is defined as a chronic immune-mediated esophageal disease with histologically eosinophil-predominant inflammation (with more than 15 eosinophils per high-power field). The diagnosis is additionally based on symptoms and endoscopic appearance.³ When investigating possible eosinophilic esophagitis, it is recommended that 2 to 4 samples be obtained from at least 2 different locations in the esophagus (eg, proximal and distal), because the inflammatory changes can be patchy.

3. What is the likely cause of this patient’s dysphagia?

• Eosinophilic esophagitis
• Achalasia
• Esophageal spasm
• Extrinsic compression
• Esophageal malignancy

Eosinophilic esophagitis causes characteristic symptoms that include difficulty swallowing, chest pain that does not respond to antisecretory therapy, and regurgitation of undigested food. As we discussed above, this patient has only 5 eosinophils per high-power field and does not meet the histologic criteria for eosinophilic esophagitis.

Achalasia has a characteristic “bird’s beak” appearance on esophagography that results from distal tapering of the esophagus to the gastroesophageal junction,¹ and this is not apparent on our patient’s study.

Review of this patient’s esophagogram also does not reveal any extrinsic compression, esophageal malignancy, or distal tapering suggesting achalasia. In light of the abrupt onset of symptoms related to both solids and liquids associated with atypical chest pain, the primary concern should be for esophageal spasm.

ONE MORE TEST

4. What study would you order next to better elucidate the cause of this patient’s esophageal disorder?

• High-resolution esophageal manometry
• Esophagogastroduodenoscopy (EGD) with endoscopic ultrasonography
• 24-hour pH and impedance testing
• Wireless motility capsule

Esophageal manometry (Figure 2) is used to evaluate the function and coordination of the muscles of the esophagus, as in disorders of esophageal motility.

High-resolution manometry is the gold standard for evaluation of esophageal motility. It is appropriate in evaluating dysphagia or noncardiac chest pain without evidence of mechanical obstruction, ulceration, or inflammation.^4,5

High-resolution manometry differs from conventional manometry in that the catheter has more sensors to measure intraluminal pressure (36 rather than the usual 7 to 12). The data are translated into pressure topography plots (Figure 3).^6,7

Updated guidelines on how to interpret the findings of high-resolution manometry are known as the Chicago 3.0 criteria.⁴ According to this system, esophageal motility disorders are grouped on the basis of lower esophageal sphincter relaxation and then further subdivided based on the character of peristalsis.

EGD with endoscopic ultrasonography would not be appropriate at this time because there is little suspicion of an extraluminal mass that needs to be investigated.

A 24-hour pH and impedance study is helpful in determining the presence of esophageal acid exposure in patients presenting with gastroesophageal reflux disease. This patient does not have symptoms of heartburn or regurgitation; therefore, this investigation would not be of value.

A wireless motility capsule would help in investigating gastric and small-bowel motility and may be useful in the future for this patient, but at this point it would provide little additional utility.

ESOPHAGEAL SPASM

Our patient underwent high-resolution esophageal manometry. The results (Figure 4) revealed a normal resting pressure in the lower esophageal sphincter and complete relaxation in all swallows. The body of the esophagus demonstrated premature contractions in 90% of swallows. Overall, these findings were consistent with the diagnosis of distal esophageal spasm.

In addition to incorporating data obtained from endoscopy, esophagography, and manometry, it is crucial to identify the patient’s predominant symptom when planning treatment. For example, is the prevailing symptom dysphagia or chest pain? Additional consideration must be given to medical, surgical, and psychiatric comorbidities.

5. Which of the following is appropriate medical therapy for esophageal spasm?

• Calcium channel blockers
• Nitrates
• Hydralazine
• Phosphodiesterase 5 (PDE5) inhibitors
• All of the above

All of these have been used to treat distal esophageal spasm as well as other hypercontractile esophageal motility disorders.^8–20

Calcium channel blockers have proven to be effective in randomized controlled trials. Diltiazem has been shown to be beneficial at doses ranging from 60 to 90 mg, as has nifedipine 10 to 20 mg 3 times daily. Although different drugs of this class tend to relax the lower esophageal sphincter to different degrees, when choosing among them in patients with hypercontractile disorders there is little concern for potentially precipitating reflux.^8–13

Nitrates, hydralazine, and PDE5 inhibitors have been effective in uncontrolled studies but have not been studied in randomized trials.^14–17

Other treatments. Patients may also benefit from neuromodulators such as trazodone and imipramine for chest pain and optimization of antisecretory therapy if they have concomitant gastroesophageal reflux disease.^18–20

Patients who have documented esophageal hypercontractility along with reflux disease confirmed by an abnormal pH study show significant improvement in their chest pain symptoms with high doses of a proton pump inhibitor (PPI). As our patient presented with chest pain and dysphagia, a dedicated pH study was not needed, and we could progress straight to manometry and a trial of a PPI.

Our patient was started on a PPI and nifedipine but developed a pruritic rash. As rash does not preclude using another medication in the same class, his treatment was changed to diltiazem 30 mg by mouth 3 times a day, and his dysphagia improved. However, he continued to experience intermittent chest pain with swallowing. After discussion of neuromodulator therapy, he declined additional pharmacologic therapy.

A NONPHARMACOLOGIC TREATMENT?

6. Which of the following would you offer this patient as a nonpharmacologic alternative for his esophageal pain?

• St. John’s wort
• Ginkgo biloba
• Ginseng
• Peppermint extract
• Eucalyptus oil

In a small, open-label study in patients with esophageal spasm, the use of 5 drops of commercially available 11% peppermint extract in 10 mL of water significantly decreased simultaneous contractions and resolved chest pain.²¹ Esophageal manometry was performed 10 minutes after the peppermint solution was consumed, and the results showed improvement in esophageal spasm. While the authors of this study did not make any formal recommendations, the findings suggest that peppermint extract should be given 10 minutes before meals.

There is no evidence for or against the use of the other nonpharmacologic treatments mentioned here.

PAIN RELIEF

7. If a pharmacologic approach were chosen, which would be the best option for pain relief in this patient?

• Oxycodone 5 mg every 8 hours
• Acetaminophen 650 mg every 8 hours
• Ibuprofen 400 mg every evening at bedtime
• Trazodone 100 mg every evening at bedtime
• Imipramine 50 mg every evening at bedtime
• Aripiprazole 5 mg by mouth every day

Trazodone would be the most appropriate of these options. Doses of 100 mg to 150 mg every evening at bedtime have been shown to significantly improve global assessment scores of pain at 6 weeks.¹⁸

Imipramine 50 mg every evening at bedtime would be another option and also has been shown to reduce chest pain.¹⁹

Even though these were the doses that were investigated, in clinical practice it is common to start at lower doses (trazodone 50 mg or imipramine 10 mg) and to then titrate every 4 weeks based on the patient’s response.

Opiates (eg, oxycodone) should be avoided, as they can cause esophageal motility disorders such as spasm or achalasia.²²

Acetaminophen and aripiprazole have not been studied exclusively for their effect on chest pain related to esophageal spasm.

RECURRENT SYMPTOMS

The patient’s dysphagia initially decreased while he was taking diltiazem 30 mg 3 times a day, but it recurred after 6 months. The dosage was increased to 60 mg 3 times a day over the course of the next year, with minimal response. (The maximum dose is 90 mg 4 times a day, but because of side effects of lightheadedness and dizziness, out patient could not tolerate more than 60 mg 3 times a day).

8. What endoscopic therapies are appropriate for patients with esophageal spasm that does not respond to medication?

• Bougie dilation
• Balloon dilation
• Onabotulinum toxin injection
• Expandable mesh stent placement
• Mucosal sclerotherapy

Onabotulinum toxin injections have been shown to improve dysphagia when given in a linear pattern.²³

Endoscopic dilation has not been shown to be beneficial in this setting, as a study found no difference in efficacy between therapeutic (54-French) and sham (24-French) bougie dilation.²⁴

Our patient received 100 units of onabotulinum toxin (10 units every centimeter in the distal 10 cm of the esophagus). Afterward, he experienced resolution of dysphagia, with only mild intermittent chest pain, which was controlled by taking peppermint extract as needed. The symptoms returned approximately 1 year later but responded to repeat endoscopy with onabotulinum toxin injections.^23,25

Peroral endoscopic myotomy

Another relatively new endoscopic treatment for esophageal motility disorders is peroral endoscopic myotomy (Figure 5). During this procedure a tiny incision is made in the esophageal mucosa, permitting the endoscope to tunnel within the lining. The smooth muscle of the distal esophagus and lower esophageal sphincter is then cut, thereby freeing either the spastic muscle (in distal esophageal spasm) or the hyperactive lower esophageal sphincter (in achalasia).^26,27

In an open trial, after undergoing peroral endoscopic myotomy for esophageal spasm and hypercontractile esophagus, 89% of patients had complete relief of dysphagia, and 92% had palliation of chest pain.²⁸ Of note, the rate of relief of dysphagia was higher for patients with achalasia (98%) than for nonachalasia patients (71%).

A 71-year-old man was referred to the gastroenterology department for evaluation of 9 months of progressive swallowing difficulties associated with epigastric and chest discomfort.

He was a previous smoker (17 pack-years), with a history of coronary artery disease, hypertension, and cervical spinal stenosis requiring decompressive laminectomy with a postoperative course complicated by episodes of aspiration.

DYSPHAGIA: OROPHARYNGEAL OR ESOPHAGEAL

Difficulty swallowing (dysphagia) can be caused by problems in the oropharynx or in the esophagus. Difficulty initiating a swallow can be thought of as oropharyngeal dysphagia, whereas the intermittent sensation of food stuck in the neck or chest is considered esophageal dysphagia.

Focused questioning can help differentiate oropharyngeal symptoms from esophageal symptoms. For example, difficulty clearing secretions or passing the food bolus beyond the mouth or frequent coughing spells while eating is consistent with oropharyngeal dysphagia and suggests a neurologic cause. Our patient, however, presented with a constellation of symptoms more suggestive of esophageal dysphagia.

When eliciting a history of esophageal symptoms, it is crucial to determine the progression of swallowing difficulty, as well as how it directly relates to eating solids or liquids, or both. Difficulty swallowing solid foods that has progressed over time to include liquids would raise concern for an obstruction such as a stricture, ring, or malignancy. On the other hand, abrupt onset of intermittent dysphagia to both solids and liquids would raise concern for a motility disorder of the esophagus. This patient presented with an abrupt onset of intermittent symptoms to both solids and liquids that was associated with substernal chest pain.

Once coronary disease was ruled out by cardiac biomarker testing, electrocardiography, and a pharmacologic stress test, our patient underwent upper endoscopy, which showed a normal esophageal mucosa without masses or obstruction and no evidence of peptic ulcer disease.

WHAT IS THE NEXT STEP?

When upper endoscopy is negative and cardiac causes and gastroesophageal reflux disease have been ruled out, an esophageal motility disorder should be considered.

1. After obstruction has been ruled out with upper endoscopy, which should be the next step in the investigation of esophageal dysphagia?

• A 24-hour pH recording
• Barium esophagography
• Modified barium swallow
• Computed tomography of the chest

Barium esophagography is the optimal fluoroscopic study to evaluate the esophageal phase of the swallow. This study requires the patient to swallow a thick barium solution and a 13-mm barium pill under video analysis. It is useful early in the investigation of esophageal dysphagia because it can potentially reveal areas of esophageal luminal narrowing not detected endoscopically, as well as detail the rate of esophageal emptying.¹

The modified barium swallow, which is performed with the assistance of a speech pathologist, is similar but only shows the oropharynx as far as the cervical esophagus. Therefore, it would be the best fluoroscopic test to assess patients with possible aspiration or oropharyngeal dysphagia, whereas barium esophagography would be the test of choice in evaluating esophageal dysmotility or mechanical obstruction.

pH testing may be helpful in diagnosing gastroesophageal reflux disease but is less helpful in the evaluation of dysphagia.

Computed tomography of the chest may be useful to evaluate for extrinsic compression of the esophagus, but it is not the best next step in the evaluation of dysphagia.

Our patient underwent barium esophagography, which revealed tertiary contractions in the mid and distal esophagus with slight narrowing of the lower cervical esophagus (Figure 1). (Primary contractions are elicited when initiating a swallow that propels the food bolus through the esophagus, while secondary contractions follow in response to esophageal distention to move all remaining esophageal contents from the thoracic esophagus. Tertiary contractions are abnormal, nonpropulsive, spontaneous contractions of the esophageal body that are initiated without swallowing.²)

EOSINOPHILIC ESOPHAGITIS

Histologic study of biopsies of the mid and distal esophagus from our patient’s upper endoscopy revealed 5 eosinophils per high-power field.

2. Does this patient meet the criteria for the diagnosis of eosinophilic esophagitis?

• Yes
• No

No. Having eosinophils in the esophagus is not enough to diagnose eosinophilic esophagitis, as eosinophils are also common in patients with gastroesophageal reflux disease.

Eosinophilic esophagitis is defined as a chronic immune-mediated esophageal disease with histologically eosinophil-predominant inflammation (with more than 15 eosinophils per high-power field). The diagnosis is additionally based on symptoms and endoscopic appearance.³ When investigating possible eosinophilic esophagitis, it is recommended that 2 to 4 samples be obtained from at least 2 different locations in the esophagus (eg, proximal and distal), because the inflammatory changes can be patchy.

3. What is the likely cause of this patient’s dysphagia?

• Eosinophilic esophagitis
• Achalasia
• Esophageal spasm
• Extrinsic compression
• Esophageal malignancy

Eosinophilic esophagitis causes characteristic symptoms that include difficulty swallowing, chest pain that does not respond to antisecretory therapy, and regurgitation of undigested food. As we discussed above, this patient has only 5 eosinophils per high-power field and does not meet the histologic criteria for eosinophilic esophagitis.

Achalasia has a characteristic “bird’s beak” appearance on esophagography that results from distal tapering of the esophagus to the gastroesophageal junction,¹ and this is not apparent on our patient’s study.

Review of this patient’s esophagogram also does not reveal any extrinsic compression, esophageal malignancy, or distal tapering suggesting achalasia. In light of the abrupt onset of symptoms related to both solids and liquids associated with atypical chest pain, the primary concern should be for esophageal spasm.

ONE MORE TEST

4. What study would you order next to better elucidate the cause of this patient’s esophageal disorder?

• High-resolution esophageal manometry
• Esophagogastroduodenoscopy (EGD) with endoscopic ultrasonography
• 24-hour pH and impedance testing
• Wireless motility capsule

Esophageal manometry (Figure 2) is used to evaluate the function and coordination of the muscles of the esophagus, as in disorders of esophageal motility.

High-resolution manometry is the gold standard for evaluation of esophageal motility. It is appropriate in evaluating dysphagia or noncardiac chest pain without evidence of mechanical obstruction, ulceration, or inflammation.^4,5

High-resolution manometry differs from conventional manometry in that the catheter has more sensors to measure intraluminal pressure (36 rather than the usual 7 to 12). The data are translated into pressure topography plots (Figure 3).^6,7

Updated guidelines on how to interpret the findings of high-resolution manometry are known as the Chicago 3.0 criteria.⁴ According to this system, esophageal motility disorders are grouped on the basis of lower esophageal sphincter relaxation and then further subdivided based on the character of peristalsis.

EGD with endoscopic ultrasonography would not be appropriate at this time because there is little suspicion of an extraluminal mass that needs to be investigated.

A 24-hour pH and impedance study is helpful in determining the presence of esophageal acid exposure in patients presenting with gastroesophageal reflux disease. This patient does not have symptoms of heartburn or regurgitation; therefore, this investigation would not be of value.

A wireless motility capsule would help in investigating gastric and small-bowel motility and may be useful in the future for this patient, but at this point it would provide little additional utility.

ESOPHAGEAL SPASM

Our patient underwent high-resolution esophageal manometry. The results (Figure 4) revealed a normal resting pressure in the lower esophageal sphincter and complete relaxation in all swallows. The body of the esophagus demonstrated premature contractions in 90% of swallows. Overall, these findings were consistent with the diagnosis of distal esophageal spasm.

In addition to incorporating data obtained from endoscopy, esophagography, and manometry, it is crucial to identify the patient’s predominant symptom when planning treatment. For example, is the prevailing symptom dysphagia or chest pain? Additional consideration must be given to medical, surgical, and psychiatric comorbidities.

5. Which of the following is appropriate medical therapy for esophageal spasm?

• Calcium channel blockers
• Nitrates
• Hydralazine
• Phosphodiesterase 5 (PDE5) inhibitors
• All of the above

All of these have been used to treat distal esophageal spasm as well as other hypercontractile esophageal motility disorders.^8–20

Calcium channel blockers have proven to be effective in randomized controlled trials. Diltiazem has been shown to be beneficial at doses ranging from 60 to 90 mg, as has nifedipine 10 to 20 mg 3 times daily. Although different drugs of this class tend to relax the lower esophageal sphincter to different degrees, when choosing among them in patients with hypercontractile disorders there is little concern for potentially precipitating reflux.^8–13

Nitrates, hydralazine, and PDE5 inhibitors have been effective in uncontrolled studies but have not been studied in randomized trials.^14–17

Other treatments. Patients may also benefit from neuromodulators such as trazodone and imipramine for chest pain and optimization of antisecretory therapy if they have concomitant gastroesophageal reflux disease.^18–20

Patients who have documented esophageal hypercontractility along with reflux disease confirmed by an abnormal pH study show significant improvement in their chest pain symptoms with high doses of a proton pump inhibitor (PPI). As our patient presented with chest pain and dysphagia, a dedicated pH study was not needed, and we could progress straight to manometry and a trial of a PPI.

Our patient was started on a PPI and nifedipine but developed a pruritic rash. As rash does not preclude using another medication in the same class, his treatment was changed to diltiazem 30 mg by mouth 3 times a day, and his dysphagia improved. However, he continued to experience intermittent chest pain with swallowing. After discussion of neuromodulator therapy, he declined additional pharmacologic therapy.

A NONPHARMACOLOGIC TREATMENT?

6. Which of the following would you offer this patient as a nonpharmacologic alternative for his esophageal pain?

• St. John’s wort
• Ginkgo biloba
• Ginseng
• Peppermint extract
• Eucalyptus oil

In a small, open-label study in patients with esophageal spasm, the use of 5 drops of commercially available 11% peppermint extract in 10 mL of water significantly decreased simultaneous contractions and resolved chest pain.²¹ Esophageal manometry was performed 10 minutes after the peppermint solution was consumed, and the results showed improvement in esophageal spasm. While the authors of this study did not make any formal recommendations, the findings suggest that peppermint extract should be given 10 minutes before meals.

There is no evidence for or against the use of the other nonpharmacologic treatments mentioned here.

PAIN RELIEF

7. If a pharmacologic approach were chosen, which would be the best option for pain relief in this patient?

• Oxycodone 5 mg every 8 hours
• Acetaminophen 650 mg every 8 hours
• Ibuprofen 400 mg every evening at bedtime
• Trazodone 100 mg every evening at bedtime
• Imipramine 50 mg every evening at bedtime
• Aripiprazole 5 mg by mouth every day

Trazodone would be the most appropriate of these options. Doses of 100 mg to 150 mg every evening at bedtime have been shown to significantly improve global assessment scores of pain at 6 weeks.¹⁸

Imipramine 50 mg every evening at bedtime would be another option and also has been shown to reduce chest pain.¹⁹

Even though these were the doses that were investigated, in clinical practice it is common to start at lower doses (trazodone 50 mg or imipramine 10 mg) and to then titrate every 4 weeks based on the patient’s response.

Opiates (eg, oxycodone) should be avoided, as they can cause esophageal motility disorders such as spasm or achalasia.²²

Acetaminophen and aripiprazole have not been studied exclusively for their effect on chest pain related to esophageal spasm.

RECURRENT SYMPTOMS

The patient’s dysphagia initially decreased while he was taking diltiazem 30 mg 3 times a day, but it recurred after 6 months. The dosage was increased to 60 mg 3 times a day over the course of the next year, with minimal response. (The maximum dose is 90 mg 4 times a day, but because of side effects of lightheadedness and dizziness, out patient could not tolerate more than 60 mg 3 times a day).

8. What endoscopic therapies are appropriate for patients with esophageal spasm that does not respond to medication?

• Bougie dilation
• Balloon dilation
• Onabotulinum toxin injection
• Expandable mesh stent placement
• Mucosal sclerotherapy

Onabotulinum toxin injections have been shown to improve dysphagia when given in a linear pattern.²³

Endoscopic dilation has not been shown to be beneficial in this setting, as a study found no difference in efficacy between therapeutic (54-French) and sham (24-French) bougie dilation.²⁴

Our patient received 100 units of onabotulinum toxin (10 units every centimeter in the distal 10 cm of the esophagus). Afterward, he experienced resolution of dysphagia, with only mild intermittent chest pain, which was controlled by taking peppermint extract as needed. The symptoms returned approximately 1 year later but responded to repeat endoscopy with onabotulinum toxin injections.^23,25

Peroral endoscopic myotomy

Another relatively new endoscopic treatment for esophageal motility disorders is peroral endoscopic myotomy (Figure 5). During this procedure a tiny incision is made in the esophageal mucosa, permitting the endoscope to tunnel within the lining. The smooth muscle of the distal esophagus and lower esophageal sphincter is then cut, thereby freeing either the spastic muscle (in distal esophageal spasm) or the hyperactive lower esophageal sphincter (in achalasia).^26,27

In an open trial, after undergoing peroral endoscopic myotomy for esophageal spasm and hypercontractile esophagus, 89% of patients had complete relief of dysphagia, and 92% had palliation of chest pain.²⁸ Of note, the rate of relief of dysphagia was higher for patients with achalasia (98%) than for nonachalasia patients (71%).

A 71-year-old man was referred to the gastroenterology department for evaluation of 9 months of progressive swallowing difficulties associated with epigastric and chest discomfort.

He was a previous smoker (17 pack-years), with a history of coronary artery disease, hypertension, and cervical spinal stenosis requiring decompressive laminectomy with a postoperative course complicated by episodes of aspiration.

DYSPHAGIA: OROPHARYNGEAL OR ESOPHAGEAL

Difficulty swallowing (dysphagia) can be caused by problems in the oropharynx or in the esophagus. Difficulty initiating a swallow can be thought of as oropharyngeal dysphagia, whereas the intermittent sensation of food stuck in the neck or chest is considered esophageal dysphagia.

Focused questioning can help differentiate oropharyngeal symptoms from esophageal symptoms. For example, difficulty clearing secretions or passing the food bolus beyond the mouth or frequent coughing spells while eating is consistent with oropharyngeal dysphagia and suggests a neurologic cause. Our patient, however, presented with a constellation of symptoms more suggestive of esophageal dysphagia.

When eliciting a history of esophageal symptoms, it is crucial to determine the progression of swallowing difficulty, as well as how it directly relates to eating solids or liquids, or both. Difficulty swallowing solid foods that has progressed over time to include liquids would raise concern for an obstruction such as a stricture, ring, or malignancy. On the other hand, abrupt onset of intermittent dysphagia to both solids and liquids would raise concern for a motility disorder of the esophagus. This patient presented with an abrupt onset of intermittent symptoms to both solids and liquids that was associated with substernal chest pain.

Once coronary disease was ruled out by cardiac biomarker testing, electrocardiography, and a pharmacologic stress test, our patient underwent upper endoscopy, which showed a normal esophageal mucosa without masses or obstruction and no evidence of peptic ulcer disease.

WHAT IS THE NEXT STEP?

When upper endoscopy is negative and cardiac causes and gastroesophageal reflux disease have been ruled out, an esophageal motility disorder should be considered.

1. After obstruction has been ruled out with upper endoscopy, which should be the next step in the investigation of esophageal dysphagia?

• A 24-hour pH recording
• Barium esophagography
• Modified barium swallow
• Computed tomography of the chest

Barium esophagography is the optimal fluoroscopic study to evaluate the esophageal phase of the swallow. This study requires the patient to swallow a thick barium solution and a 13-mm barium pill under video analysis. It is useful early in the investigation of esophageal dysphagia because it can potentially reveal areas of esophageal luminal narrowing not detected endoscopically, as well as detail the rate of esophageal emptying.¹

The modified barium swallow, which is performed with the assistance of a speech pathologist, is similar but only shows the oropharynx as far as the cervical esophagus. Therefore, it would be the best fluoroscopic test to assess patients with possible aspiration or oropharyngeal dysphagia, whereas barium esophagography would be the test of choice in evaluating esophageal dysmotility or mechanical obstruction.

pH testing may be helpful in diagnosing gastroesophageal reflux disease but is less helpful in the evaluation of dysphagia.

Computed tomography of the chest may be useful to evaluate for extrinsic compression of the esophagus, but it is not the best next step in the evaluation of dysphagia.

Our patient underwent barium esophagography, which revealed tertiary contractions in the mid and distal esophagus with slight narrowing of the lower cervical esophagus (Figure 1). (Primary contractions are elicited when initiating a swallow that propels the food bolus through the esophagus, while secondary contractions follow in response to esophageal distention to move all remaining esophageal contents from the thoracic esophagus. Tertiary contractions are abnormal, nonpropulsive, spontaneous contractions of the esophageal body that are initiated without swallowing.²)

EOSINOPHILIC ESOPHAGITIS

Histologic study of biopsies of the mid and distal esophagus from our patient’s upper endoscopy revealed 5 eosinophils per high-power field.

2. Does this patient meet the criteria for the diagnosis of eosinophilic esophagitis?

• Yes
• No

No. Having eosinophils in the esophagus is not enough to diagnose eosinophilic esophagitis, as eosinophils are also common in patients with gastroesophageal reflux disease.

Eosinophilic esophagitis is defined as a chronic immune-mediated esophageal disease with histologically eosinophil-predominant inflammation (with more than 15 eosinophils per high-power field). The diagnosis is additionally based on symptoms and endoscopic appearance.³ When investigating possible eosinophilic esophagitis, it is recommended that 2 to 4 samples be obtained from at least 2 different locations in the esophagus (eg, proximal and distal), because the inflammatory changes can be patchy.

3. What is the likely cause of this patient’s dysphagia?

• Eosinophilic esophagitis
• Achalasia
• Esophageal spasm
• Extrinsic compression
• Esophageal malignancy

Eosinophilic esophagitis causes characteristic symptoms that include difficulty swallowing, chest pain that does not respond to antisecretory therapy, and regurgitation of undigested food. As we discussed above, this patient has only 5 eosinophils per high-power field and does not meet the histologic criteria for eosinophilic esophagitis.