The epidemic of type 2 diabetes mellitus (T2DM) presents the clinician as well as society with complex challenges in designing both prevention and treatment strategies for T2DM. T2DM is a worldwide epidemic with a global prevalence in 2009 of 285 million people, expected to increase to 435 million by 2030 [1]. The Centers for Disease Control and Prevention (CDC) estimates that in 2007 nearly 24 million people in the USA have diabetes [2], mostly T2DM. At least 57 million people have pre-diabetes (impaired fasting glucose or impaired glucose tolerance), providing an expanding pool of new patients with T2DM. Current prevalence represents a 90% increase in the new diagnosis of diabetes over the last decade [3] and more rapid increases in many parts of the world. This increasing prevalence of diabetes relates both to environmental and genetic factors, each of which may influence insulin sensitivity (e.g. insulin resistance) and insulin secretion capacity. Insulin resistance begins in early life and is fueled by obesity and a sedentary lifestyle which contribute to alterations in glucose homeostasis and to abnormal lipid and protein metabolism.

Relative insulin deficiency is the defining metabolic difference between obesity and the development of hyperglycemia with progression to T2DM. An inexorable progression of diabetes appears related to worsening insulin deficiency [4,5]. Most people therefore have gradually increasing needs for additional therapy. Combinations of therapy such as adopting a therapeutic lifestyle change, as well as oral and injectable medicines, are required to keep the glycemia under control. The failure to advance therapy at an early sign of therapeutic failure and, in particular, a reluctance to advance to insulin underlies the less than optimum control for many patients with diabetes.

Treatment goals for patients with diabetes have changed considerably over the past two decades. The Diabetes Control and Complications Trial (DCCT) [7] confirmed the concept of the relationship of glycemic control to microvascular complications in type 1 diabetes. Similar results were found for T2DM in the landmark UK Prospective Diabetes Study (UKPDS) [8,9]. Long-term analysis of subjects participating in these trials at 10 years following a period of improved glycemic control found cardiovascular risk reduction with insulin, metformin and sulfonylureas [10,11] and for T2DM also a mortality advantage.

In assigning patients to a treatment plan, consideration must be given to treatment goals. The ADA continues to recommend HbA_1c of <7% (<53 mmol/mol) as a general glycemic goal and individualization is recommended. For properly selected patients, either less tight control or alternatively more rigorous normalization of glycemia down to 6% (42 mmol/mol) HbA_1c can be recommended, if the latter can be achieved without hypoglycemia problems. Using similar data sources, the International Diabetes Federation (IDF) and American Association of Clinical Endocrinologists (AACE) suggest ≤6.5% (48 mmol/mol) as a general goal. All authorities recommend the need for individualization of goals based on co-morbidities and vulnerability to hypoglycemia among other criteria. The ACCORD trial hypothesized that close to normal glucose control (HbA_1c 6% or less [<42 mmol/mol]) would improve cardiovascular endpoints. Recent analysis of the ACCORD trial [12] cautions the provider against attempting to normalize glycemia in those with characteristics similar to this patient cohort (e.g. older and those with existing cardiovascular disease), as the intensively treated group had a 22% increase in mortality primarily from fatal cardiovascular events; very possibly, although unproven, this may have been as a result of hypoglycemia. This poses a dilemma to the practitioner: how best to normalize glycemia but avoid hypoglycemia. This practical consideration suggests newer therapies with a relatively low hypoglycemia risk should be considered for certain patient groups. Additionally, strategies to recognize and prevent hypoglycemia risk in patients at high risk should be adopted routinely. These include the need for self-monitored blood glucose (SMBG) tests, good communication and education about how to treat and adjust therapy should hypoglycemia become frequent or severe, especially if hypoglycemia unawareness occurs.

Ultimately, patients with T2DM require progression of therapy from combination oral therapy to combination injectable therapy (basal bolus insulin regimens, incretin insulin combinations). Initiating combination therapy early minimizes side effects while maximizing clinical effectiveness, decreasing pill counts and cost while facilitating compliance. New ACCE guidelines use all drugs and tailor their number and starting insulin to the baseline HbA_1c [13].

Lifestyle management is one of the most challenging barriers to successful diabetes control. Deriving recommendations for eating, physical activity and minimizing psychologic stressors is frequently frustrating for both the provider and patient [14]. The assumption that complex patterns of lifestyle behavior can be altered by dispensing general advice in the course of a busy medical visit is flawed, but unfortunately a reality for many providers. It should be replaced by an agreement that the patient is responsible for assisting in their disease management and is the only source of key information needed to manage their illness [15]. One strategy is to make patients more aware of their own lifestyle patterns. Developing a baseline for eating and exercise (type, frequency, duration, barriers) is as integral as obtaining initial laboratory testing in patients with T2DM to compile a treatment plan based on physiology as well as the psychology of the patient. Obtaining a narrow focus on a few behavioral objectives can increase success. Approaches we often use for patients are:

• Keep an eating behavior diary over the next week; and
  
• Wear a pedometer and keep track of your baseline steps for the next week.

The information almost always contains some surprises for patients but it helps them to identify personal behaviors and to focus and motivate them to take the lead in setting meaningful objectives.

The role of the physician is critical in emphasizing the importance of lifestyle change to the patient, initiating a process of working on lifestyle change, and reinforcing, and refining the education to achieve incremental obtainable objectives. Use of motivational interviewing and assessment of readiness to change are important aspects of evaluation of the patient by both the medical provider and diabetes educator. It is our practice to recommend strongly diabetes education for every patient who is willing to do so and to reinforce the recommendation for lifestyle change at any point when additional therapy needs occur. A useful acronym that we have used to help trainees remember how to assist patients in making lifestyle change is FIRM [15]. This stands for negotiating Few changes, typically 1–3 at any clinical visit; those changes should be Individualized by the patient’s selection of what aspects of behavior change to embark on; the changes should be Realistic and therefore likely to succeed by setting moderate, achievable goals within a specific timeframe; and be Measureable and monitored through patient record-keeping to be shared with the provider or diabetes educator at the next visit.

Because the vast majority of people with T2DM are overweight or obese, a serious consideration is the use of weight loss and increasing physical activity strategies as a useful treatment to gain control of glycemia and favorably influence multiple cardiovascular risk factors as well, 16,17]. Prevention and/or treatment plans for T2DM could also legitimately consider use of pharmacotherapy for obesity. Successful weight loss of 10% or less has repeatedly been shown to improve glycemic control (HbA_1c reductions 0.30–0.5% [3–6 mmol/mol] and sometimes much more) as well as favorable improvements in lipids, hypertension as well as other cardiac risk factors. Currently, three drugs (phentermine, orlistat and sibutramine) are approved for management of obesity in the USA. Rimonabant was approved for use in several countries outside the USA, but recent withdrawal of support by the European Medicines Agency has ended its possibility of wider use. Other drugs of the endocannabinoid receptor blocker class have also been discontinued from clinical development. Phentermine is only approved for short-term use and is the least well-studied and is less commonly used and not generally recommended.

Orlistat is currently approved for use in the management of obesity. Orlistat functions as a gastric and pancreatic lipase inhibitor. When given at a dosage of 120 mg three times daily with meals, approximately 20–30% of ingested fat is passed into the feces. Large amounts of fat in the gastrointestinal tract result in the most difficult side effect to manage, namely anal leakage of oil to post-prandial diarrhea which may be abrupt. Attention to this expected effect as well as dietary counseling or use of daily or twice daily dosing can minimize patient’s gastrointestinal accidents and may enhance patient adherence to low fat dietary regimens. No long-term malabsorption of fat-soluble vitamins has been seen but replacement with a multivitamin including vitamin D is recommended during therapy. A lower dosage (60 mg) over-the-counter form is now available.

Orlistat has been critically studied both in the prevention [18] and in the treatment of T2DM in patients receiving metformin or insulin [19,20]. In the XENDOS trial, 3305 obese subjects at risk for T2DM were randomized to therapeutic lifestyle change or therapeutic lifestyle change plus orlistat 120 mg three times daily over a study period of 4 years. The subset of subjects with impaired glucose tolerance at entry receiving orlistat enjoyed a 37% relative risk reduction in development of T2DM at study end. Favorable lipid effects were also seen.

In obese subjects with T2DM failing metformin monotherapy [19], the addition of orlistat to the treatment plan lowered HbA_1c by 0.35% (4 mmol/mol) and improved lipid as well as blood pressure control. Similarly, benefits were also seen in insulin-treated patients given orlistat for 1 year [20] with improved glycemia (HbA_1c –0.62 ± 0.08 for orlistat-treated vs –0.27 ± 0.08% given placebo; P < 0.002, 7 vs 3 mmol/mol) and improved lipids despite lowered diabetes medicine doses.

Therapies for T2DM can be divided into drugs facilitating supply of endogenous insulin (secretagogues, incretins) or those enhancing insulin actions (biguanides, thiazolidinediones, incretins, α-glucosidase inhibitors, amylin analogs, colesevelam). Successful treatment of T2DM combines approaches to lifestyle modification frequently in conjunction with use of pharmacologic therapy.

We conclude from data found in the UKPDS that insulin secretory loss starts on average 10 years before the formal diagnosis of T2DM, followed by insulin deficiency on average 10–12 years after diagnosis [8]. The dual aspects of T2DM, insulin resistance and insulin deficiency, are important factors in therapy selection and the subsequent response to therapy. This chapter provides an update on combinations of therapies for T2DM, presents evidence and opinions on sequences of medications, reviews new insights into diabetes and obesity treatments, and briefly describes some new medications emerging for the treatment of T2DM.

In the context of at least two major physiologic defects in T2DM, insulin resistance and progressive insulin secretory failure, combining treatments with complementary actions can be a logical approach. Several distinct advantages of combining agents can be distinguished.

Efficacy

The antihyperglycemic effect of metformin is largely brought about by suppression of hepatic glucose release, thus reducing insulin resistance in the liver. Metformin may accomplish this by activation of the enzyme adenosine 5’-monophosphate-activated protein kinase (AMPK), which acts as a “fuel sensor.” Although not approved by the FDA for the treatment of the metabolic syndrome or intermediate hyperglycemia, some clinicians have taken advantage of the insights from the Diabetes Prevention Program (DPP) [35]; in patients with impaired glucose tolerance, there was a delay in the onset of T2DM through the use of intensive lifestyle modification (58% reduction), metformin (31% reduction) and troglitazone (early DPP data suggest a benefit, but this study arm was discontinued because of the hepatotoxicity observed). Thus, in patients with pre-diabetes or the metabolic syndrome, metformin is sometimes used in combination with intensive lifestyle modification for prevention of T2DM.

Metformin is usually the preferred initial treatment selection in most T2DM. Data, again from the UKPDS, demonstrate that metformin as monotherapy in overweight patients reduced mortality, myocardial infarction and weight gain [9]. Long-term follow-up from the UKPDS suggests this cardiovascular and mortality benefit persists [11]. The use of metformin in combination with secretagogues in T2DM is very common and two combination pills are available. One formulation combines metformin with glyburide, and the other combines metformin with glipizide. Combination medications are also available for metformin with thiazolidinediones, repaglinide and sitagliptin. The combination medications do not include the extended-release versions of metformin, and this may limit gastrointestinal tolerance for a few patients.

Thiazolidinediones (TZDs) are insulin-sensitizing agents that act as ligands of the nuclear transcription factor peroxisome proliferator-activated receptor γ (PPAR-γ) [36]. In so doing, they improve not only glycemia but may ameliorate dyslipidemia, inflammation and hypercoagulability associated with insulin resistance [37]. As a result, there is particular interest in their potential cardiovascular benefits in T2DM. The pleiotropic effects of TZDs beyond glycemic control lie in the role of their ligand, PPAR-γ, on lipid metabolism and inflammatory pathways. PPARs are members of a nuclear receptor superfamily that regulates gene expression in response to ligand binding. There are three known PPARs: α, δ (sometimes called β) and γ. PPAR-α is found in the liver, the heart, muscle and vascular walls, and binds fibrates, with subsequent free fatty acid oxidation, reduced triglycerides, improved high-density lipoprotein (HDL) cholesterol and a decrease in inflammation [37]. PPAR-γ is most abundant in adipose tissue, but also is in β-cells of pancreatic islets, vascular endothelium and macrophages. It regulates gene expression, influencing adipocyte differentiation, fatty acid uptake and storage, and glucose uptake [36,37].

Rosiglitazone and pioglitazone are both FDA approved for monotherapy and in combination with metformin and sulfonylureas [38–41]. Because of concerns about cardiovascular risk, rosiglitazone is no longer recommended with insulin or nitrate therapy. Pioglitazone is approved for use with insulin. A TZD predecessor, troglitazone, was FDA approved in 1997 but withdrawn in 2000 because of reports of fatal hepatotoxicity, which is not seen with current TZDs. In a meta-analysis of 13 randomized clinical trials, less than 0.3% of patients on pioglitazone or rosiglitazone had alanine aminotransferase (ALT) levels greater than three times the upper limit of normal, compared to nearly 2% of patients on troglitazone [42]. Nonetheless, it is recommended that liver function tests be performed prior to initiation of therapy and periodically thereafter. Of interest, recent reports suggest that TZD therapy may improve elevated transaminase values and even liver histology in patients with T2DM and non-alcoholic fatty liver disease or non-alcoholic steatohepatitis [43].

Both pioglitazone and rosiglitazone effectively lower HbA_1c, up to about 1–1.5% (11–16 mmol/mol), with maximum doses in monotherapy [44] but less on average. Glycemic control is improved on average more than with nateglinide or α-glucosidase inhibitors, but less than with sulfonylureas or metformin in unselected patients, partly because of a heterogeneous response; some patients have a small response, but others have a robust response. When TZDs are added to sulfonylureas, metformin or insulin, an additional 0.8–1.5% (9–16 mmol/mol) HbA_1c decline can generally be achieved [45]. A combination form of metformin and rosiglitazone is currently available. A combination pill of pioglitazone and metformin is now marketed. As with the sulfonylurea combination pills, the immediate-release form of metformin is used in these formulations. For very insulin-resistant patients (usually with high BMI, marked central obesity and hypertriglyceridemia) with preserved insulin secretion (shorter duration of diabetes), use of metformin with a TZD can be very effective. These same characteristics probably identify good TZD monotherapy responders. Combination pills of metformin and rosiglitazone or pioglitazone may be preferable for some patients because there are fewer pills to take or there is lower co-payment from insurance in some countries. TZD-secretagogue combinations are effective in insulin-resistant patients with moderate insulin secretory reserve. Combination pills are available, both for pioglitazone (with glimepiride) and rosiglitazone (with glimepiride).