Metformin, which lowers glucose by reducing hepatic glucose production and increasing intestinal glucose uptake, has long been the recommended first-line agent in the treatment of type 2 diabetes. Metformin effectively reduces glycated hemoglobin (HbA1c) levels, may promote modest weight loss, and does not lead to hypoglycemia when used as monotherapy. It is also inexpensive and has a generally favorable safety profile. Metformin may also reduce the risk of long-term cardiovascular events when given as first-line therapy to patients with new-onset type 2 diabetes.

The United Kingdom Prospective Diabetes Study (UKPDS) was an early landmark trial in type 2 diabetes that investigated the effect of intensive glucose lowering on microvascular and macrovascular outcomes. The trial was begun in 1977 and the results were published in 1998. In this trial, 3867 patients with newly diagnosed type 2 diabetes (median age 54) were randomized to intensive treatment with sulfonylureas (chlorpropamide, glibenclamide [glyburide in the United States], or glipizide) or with insulin, versus conventional therapy with diet alone. A subgroup of UKPDS patients who were overweight (>120% ideal body weight) were randomized either to intensive therapy with metformin ( n = 342, median HbA1c 7.4%) or conventional diet therapy ( n = 411, median HbA1c 8%). In the overall trial, tighter glycemic control (median HbA1c level 7% in the intensive group vs. 7.9% in the conventional group) reduced the risk for a broad, diabetes-related endpoint by 12% ( P =.03) over 10 years of follow-up. This reduction in risk was primarily driven by a 25% risk reduction in microvascular events; no reduction in the risk of mortality was found, and a numeric reduction in the risk of MI did not reach statistical significance (−16%, P =.052). Further, an unfavorable point estimate for stroke with intensive therapy was identified but the increase was not statistically significant (RR [relative risk], 1.14; 95% CI [confidence interval], 0.70–1.84). However, in the metformin-treated subset of patients, intensive management with metformin was associated with a 32% reduction in any diabetes-related endpoint ( P =.002) and a 36% reduction in mortality ( P =.011) that exceeded the reductions in risk seen with sulfonylurea or insulin therapy. In summary, the UKPDS trial established that intensive glucose control reduces the risk of microvascular complications in patients with newly diagnosed type 2 diabetes, but also suggested that only overweight patients treated with metformin therapy will derive significant macrovascular benefits within the first 10 years of treatment.

After the UKPDS trial was completed, study participants and their clinicians were advised to pursue appropriate glycemic control, and patients were returned to community or hospital-based diabetes care according to their clinical needs without any attempts to maintain previously randomized therapies. In the 10-year posttrial monitoring study of patients who survived to the end of the UKDPS trial, HbA1c levels were no longer different between the original intensive and conventional arms (approximately 8% at the end of the posttrial monitoring period). In the sulfonylurea-insulin group, RR reductions for diabetes-related endpoints persisted, whereas significant risk reductions for MI (15%, P = .010) and mortality (13%, P = .007) emerged over time. In the metformin group, RR reductions were shown for any diabetes-related endpoint, MI (33%, P = .005), and mortality (27%, P = .002). These observations suggest a modest but sustained effect of intensive glucose lowering on cardiovascular events, but only after many years of follow-up.

Despite the many advantages of metformin, data with respect to cardiovascular risk reduction with metformin are not entirely consistent. In addition to the UKPDS substudy, only one prior randomized controlled trial was conducted on this subject. In the trial, 390 patients treated with insulin were randomized to receive either metformin or a placebo as add-on therapy. The primary endpoint, an aggregate of microvascular and macrovascular outcomes, did not differ between the two groups after 4.3 years of follow-up (HR [hazard ratio], 0.92; P = .33), but there was a significant reduction in the secondary endpoint of macrovascular events (HR, 0.60; P = .04). In addition, metformin use was associated with beneficial effects on body weight and insulin requirements.

In observational studies, metformin use (either as monotherapy or in combination with another oral agent) has been associated with reduced mortality, cardiovascular deaths, and cardiovascular events. Because metformin is generally the preferred initial agent for diabetes treatment and remains contraindicated in patients with advanced chronic kidney disease, patients who are not treated with this medication in an observational study may differ in important ways from those who are. These analyses either adjusted for potential confounders or matched patient populations for the propensity to be prescribed metformin versus another medication (usually a sulfonylurea). However, these investigations were observational in nature, so unmeasured factors may potentially still have contributed to the differences in outcomes.

Alongside the evidence that supports the safety and effectiveness of metformin, there are data that provide a less reassuring picture. Although the UKPDS substudy showed benefits of metformin in overweight participants, the trial also reported an increased death rate in nonoverweight patients who took metformin and a sulfonylurea compared with those who took a sulfonylurea alone (RR, 1.60, P = .04). Combined analysis of the two UKPDS studies did not reveal an increased risk for mortality in patients treated with this combination, and the increased mortality in the UKPDS substudy has not been fully explained. In a subsequent meta-analysis of 13 randomized controlled trials involving more than 13,000 patients with type 2 diabetes, compared with other treatments, metformin therapy had no significant effect on the risk for mortality (RR 0.99 with wide 95% CIs that could not exclude a 25% reduction or 31% increase in risk), cardiovascular mortality (RR, 1.05), or rates of MI (RR 0.90 with 95% CIs that could not exclude 26% risk reduction or 9% harm). Similar findings were reported in a meta-analysis that specifically examined the effects of metformin with insulin compared with insulin alone in 23 trials with over 2000 participants. The study found that metformin added to insulin did not significantly change mortality risk (RR, 1.30, with 95% CI, 0.57–2.99) or cardiovascular mortality risk (RR, 1.70, with 95% CI, 0.35–8.30) but provided little reassurance with regard to each of these endpoints given the wide CIs.

Interestingly, analyses of observational studies including approximately 34,500 patients with heart failure (HF) and type 2 diabetes suggest a benefit to the use of metformin for glycemic control compared with older antihyperglycemic therapy. Patients with HF receiving metformin ( n = 6624) had a 20% lower adjusted death rate compared with other antihyperglycemic medications (primarily sulfonylureas): 23% versus 37%; pooled adjusted HR, 0.80; 95% CI, 0.74–0.87 with no increase in risk of lactic acidosis. This suggests that among the older classes of antihyperglycemic medications, metformin represents a safer choice than sulfonylureas for patients with type 2 diabetes and HF.

Sulfonylureas, the most commonly prescribed insulin secretagogues, are the oldest oral agent class for treatment of hyperglycemia. Whereas sulfonylureas are effective at glucose lowering, they increase the risk of hypoglycemia, are associated with a modest weight gain, and may provide less durable glycemic control compared with other antihyperglycemic classes.

Hypoglycemia, which commonly occurs during treatment with sulfonylureas, other insulin secretagogues, and insulin, has been associated with adverse events in patients with diabetes including increased mortality, higher risk of dementia, falls, fall-related fractures, cardiovascular events, and poor health-related quality of life. The relationship between hypoglycemia and cardiovascular events has been of particular clinical concern. There are a number of plausible mechanisms through which acute hypoglycemia may trigger ischemia, arrhythmia, and cardiovascular events. Hypoglycemia increases levels of counterregulatory hormones, such as epinephrine and norepinephrine, which may induce increased cardiac rate and/or contractility, heightening myocardial oxygen consumption, while also precipitating vasoconstriction and platelet aggregation. Acute hypoglycemia in the presence of hypokalemia prolongs cardiac repolarization and increases the QT interval, favoring a proarrhythmic state. One study of patients with type 1 and type 2 diabetes who presented to the hospital with severe hypoglycemia documented frequent hypokalemia, QT prolongation, and severe hypertension during the hypoglycemic events.

Although animal studies have verified the effect of hypoglycemia on myocardial ischemic injury, the data in humans are less clear, and debates continue on whether hypoglycemia is a mediator or merely a marker of such adverse outcomes—that is, the propensity toward hypoglycemia may simply identify sicker individuals who have lost the ability to counterregulate. Other risk factors for hypoglycemia include intensive glucose lowering, older age, lower health literacy level, longer duration of diabetes, renal impairment, polypharmacy, baseline cognitive impairment, and recent admission to the hospital within 30 days. Notably, analyses of cardiovascular outcomes trials data have suggested that the relationship between severe hypoglycemia and cardiovascular events may be bidirectional, and each type of event indicative of a common at-risk type 2 diabetes frail patient phenotype.

Questions about the cardiovascular safety of sulfonylureas date back to 1970 when the University Group Diabetes Program (UGDP) trial reported an increased risk of cardiovascular death associated with the use of tolbutamide compared with a placebo or insulin. Subsequently, the FDA mandated a boxed warning to that effect for all sulfonylureas that remains in every sulfonylurea label to this day. Sulfonylureas inhibit the adenosine triphosphate–dependent potassium channels that are present within cardiomyocytes and coronary vascular endothelial cells, and it has been postulated that the presence of sulfonylureas at the time of an acute coronary event prevent adequate coronary vasodilation and thus may result in a larger area of myocardial damage. However, the risk of cardiovascular death noted by the UGDP group was not supported by the subsequent UKPDS study, which showed no difference in cardiovascular risk between the use of sulfonylureas (chlorpropamide, glibenclamide [aka glyburide], or glipizide) or insulin therapy (but a benefit with the use of metformin, as noted earlier ). The large A Diabetes Outcome Prevention Trial (ADOPT) compared the glycemic effectiveness of metformin, rosiglitazone, or glyburide therapy, but again identified no between-group difference with respect to cardiovascular events (which were not frequent and were collected as adverse events during the trial) after 4 years of treatment.

The Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial assessed the effects of intensive glucose lowering in patients with type 2 diabetes at high cardiovascular risk, preferentially using the sulfonylurea gliclazide in its intensive glucose control strategy. In ADVANCE, 11,140 patients with type 2 diabetes (mean age 66, median baseline HbA1c 7.2%, 32% with history of major macrovascular disease) were randomized to intensive therapy (primarily with the sulfonylurea gliclazide, targeting HbA1c ≤6.5%, with mean achieved HbA1c 6.5%) or to standard therapy (HbA1c goal according to local guidelines, mean achieved HbA1c 7.3%). After a median follow-up of 5 years, there was a reduction in the risk of the primary outcome of the study, which was a composite of microvascular and macrovascular events (HR, 0.90; P = .01); the reduction in risk was almost entirely attributable to effects on intermediate markers of nephropathy. There was no significant between-group difference with respect to the risk of major cardiovascular events (HR, 0.94; P = .32) in ADVANCE, and no increase in mortality (HR, 0.93; P = .28).

Most recently, the cardiovascular effects of sulfonylurea therapy have been assessed in the Cardiovascular Outcome Study of Linagliptin versus Glimepiride in Type 2 Diabetes (CAROLINA), which assessed the cardiovascular impact of treatment with the dipeptidyl-peptidase 4 inhibitor linagliptin to the sulfonylurea glimepiride in 6033 patients with type 2 diabetes and established or multiple risk factors for atherosclerotic cardiovascular disease. At a median 6.3 years of follow-up, the treatment groups had similar rates of the primary composite outcome (cardiovascular death, nonfatal MI, or nonfatal stroke); 11.8% with linagliptin versus 12.0% with glimepiride. The upper limit of the 95.47% CI of the HR was 1.14, which met prespecified criteria for noninferiority. The CAROLINA findings indirectly support the cardiovascular safety of glimepiride therapy, as linagliptin itself did not increase the risk of major adverse cardiovascular events compared with a placebo in the Cardiovascular and Renal Microvascular Outcome Study with Linagliptin (CARMELINA).

The Nateglinide and Valsartan in Impaired Glucose Tolerance Outcomes Research (NAVIGATOR) trial assessed the impact of postprandial glucose-lowering with a short-acting insulin secretagogue, nateglinide, in addition to lifestyle modification, in patients with impaired glucose tolerance. The trial randomized 9306 persons (mean age 64) with baseline HbA1c 5.8% and high risk for cardiovascular disease (24% had a prior history of cardiovascular events) to nateglinide or placebo. Over a median 5 years of follow-up, nateglinide did not reduce the occurrence of the three coprimary outcomes—incident diabetes (HR, 1.07; P = .05), cardiovascular outcomes (death from cardiovascular causes, nonfatal MI, nonfatal stroke, or hospitalization for HF; HR, 0.94; P = .43), or the extended cardiovascular outcome (which in addition included unstable angina or arterial revascularization; HR, 0.93; P = .16). Thus use of an insulin secretagogue cannot be recommended in patients with prediabetes to reduce the risks of either diabetes or cardiovascular events.

As type 2 diabetes progresses, beta-cell function declines and the effectiveness of oral agents in maintaining glucose control diminishes. However, endogenous insulin secretion is not entirely lost in type 2 diabetes, and strategies for insulin use can be less complex than those used in the treatment of type 1 diabetes. The main side effects of insulin use are weight gain and hypoglycemia. As insulin has been used in clinical practice for decades and is a naturally occurring peptide hormone, fewer concerns have been previously raised about the cardiovascular safety of such therapy. However, insulin may obviously induce hypoglycemia, which may in turn promote adrenergic discharges and arrhythmia, promotes fluid retention, and may result in hypokalemia (especially when administered intravenously) as a result of potassium shifts from the extracellular to intracellular space.

Concerns about the cardiovascular safety of insulin use in type 2 diabetes have been raised, primarily based on observational studies showing an increased risk of overall and cardiovascular mortality, and cardiovascular events in patients treated with insulin compared with other antihyperglycemic agents. In one such analysis, data from over 84,000 patients with type 2 diabetes in the UK General Practice Research Database were used to determine the risk of the primary outcome (first major adverse cardiac event, first cancer, or mortality) associated with five different glucose-lowering regimens (metformin monotherapy, sulfonylurea monotherapy, insulin monotherapy, metformin plus sulfonylurea, and insulin plus metformin). When compared with all other regimens, insulin monotherapy was associated with an increased risk of the primary outcome and all-cause mortality. The observational nature of the study does not permit conclusions about the causal nature of these associations; the risk of harms may be mediated by insulin itself or by the clinical characteristics of patients who require and are prescribed insulin. Indeed, there were important differences in baseline characteristics among treatment groups; despite adjustment for a variety of variables, residual confounding remains an important limitation.

Although few trials have specifically focused on the effectiveness and safety of insulin for prevention of cardiovascular events, several trials have tested the use of insulin during and after an acute MI for secondary prevention of cardiovascular events and mortality. The first Diabetes Mellitus, Insulin Glucose Infusion in Acute Myocardial Infarction (DIGAMI) study showed significant reduction in mortality (absolute reduction of 11%) associated with the use of insulin-glucose infusion during admission and subcutaneous insulin for glycemic control after hospitalization for acute MI compared with standard therapy in patients with diabetes. Because it was unclear whether the mortality benefit was the result of inpatient or outpatient glycemic control in DIGAMI, the DIGAMI-2 study tested three different strategies, including combinations of insulin-based inpatient and outpatient glycemic control regimens. However, the study struggled with patient recruitment and did not achieve differences in glycemic control among the three groups. There were no differences in mortality among the three groups. However, in the post hoc analysis of the DIGAMI-2 trial, insulin treatment was actually associated with an increased risk for nonfatal cardiovascular events, but not mortality. Therefore, although insulin treatment is a particularly effective means of achieving glycemic control, it cannot be recommended as an intervention to reduce cardiovascular risk in the setting of an acute coronary event.

Cardiovascular outcomes with the use of two different, long-acting basal insulin analogues were compared in the Trial Comparing the Cardiovascular Safety of Insulin Degludec versus Insulin Glargine in Patients with Type 2 Diabetes at High Risk of Cardiovascular Events (DEVOTE). DEVOTE enrolled 7637 patients with long-standing type 2 diabetes, 82.5% of whom also had cardiovascular and/or kidney disease, who were then randomly assigned to once-daily treatment with either the newer insulin analogue degludec or insulin glargine. At 24 months of follow-up, severe hypoglycemia was significantly less frequent in the degludec-treated group. However, rates of the primary outcome of death from cardiovascular causes, nonfatal MI, or nonfatal stroke were similar: 8.5% in the degludec group and 9.3% in the glargine group (HR, 0.91; 95% CI, 0.78–1.06; P <.001 for noninferiority). Despite an association noted between severe hypoglycemia and subsequent HF hospitalization (HHF), there were no between-group differences in rates of hospitalization for HF. As DEVOTE assessed the cardiovascular effects of two different types of insulin, the trial findings are not useful in comparing the safety of insulin to other, noninsulin antihyperglycemic therapy.

Medications that lower glucose through effect on the incretin system have been relatively recently introduced into the type 2 diabetes pharmacopeia. These agents either mimic the effects of or increase the circulating concentrations of the endogenous incretin hormone glucagon-like peptide 1 (GLP-1), which include stimulation of pancreatic insulin secretion in a glucose-dependent fashion, suppression of pancreatic glucagon output, slowed gastric emptying, and decreases in appetite through central mechanisms. The oral dipeptidyl peptidase 4 (DPP-4) inhibitors slow the metabolism and inactivation of endogenously produced GLP-1, resulting in modest decreases in HbA1c with little impact on body weight. The main advantages of the GLP-1 receptor agonists (GLP-1RAs), which most closely mimic the effects of endogenous GLP-1, are their potency in both weight loss and glucose lowering. A potentially limiting side effect of GLP-1RA therapy may be nausea and vomiting, particularly early in the course of treatment and with dose uptitration, and both classes have been associated with an increased risk of pancreatitis and/or biliary disease. Typically, neither of these incretin-based classes causes hypoglycemia as monotherapy as their hypoglycemic effects are glucose dependent.

DPP-4 inhibitors (DPP4i), newer oral agents that have been studied in numerous large outcomes trials to date, appear to neither increase nor decrease the risk of major adverse cardiovascular events compared with placebo. These trials include assessment of the cardiovascular safety of saxagliptin in the Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction 53 (SAVOR-TIMI 53) ; alogliptin in the Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care (EXAMINE) ; sitagliptin in the Trial Evaluating Cardiovascular Outcomes with Sitagliptin (TECOS) ; and linagliptin in CARMELINA and the previously described CAROLINA trial. The inclusion criteria, patient populations, study interventions, and primary outcomes of these trials are summarized in Table 14.1 . In all of the trials other than CAROLINA, the effects of DPP4i therapy have been compared with placebo in the context of otherwise usual care.

Unexpectedly, a significantly increased risk of hospitalization for HF was noted with saxagliptin treatment in the SAVOR-TIMI 53 trial. In that trial, 16,492 patients with type 2 diabetes who had risk factors or a history of cardiovascular events were assigned to receive saxagliptin or placebo in addition to their existing diabetes medication regimen. After a median of 2.1 years, there was no difference in the primary endpoint (MI, ischemic stroke, or cardiovascular death) between the two groups (HR, 1.0; 95% CI, 0.89–1.12). However, more saxagliptin-treated patients were hospitalized for HF (3.5% vs. 2.8%; HR, 1.27 [1.07–1.51]; P = .007) compared with those assigned to placebo. This increase in HF risk has not been clearly explained; however, participants at greatest risk for saxagliptin-associated HF were those with elevated levels of natriuretic peptides, prior HF, or chronic kidney disease. In the EXAMINE trial, there were also numerically, but not statistically, more HHF events in the alogliptin group compared with a placebo (106 vs. 89; HR, 1.19; P = .22). As the SAVOR-TIMI findings have not been replicated in other DPP4i cardiovascular outcomes trials, it is unclear whether this increase in HHF risk represents a true within-class difference or is perhaps due to play of chance. Notably, the more robust, prospectively planned HF analyses of TECOS, CAROLINA, and CARMELINA outcomes demonstrated no increase in risk of HHF with use of sitagliptin or linagliptin in those trials. A recent meta-analysis including data from 182 DPP4i trials identified no significant increase in risk of incident HF for the class, with the exception of saxagliptin. However, the saxagliptin effects were primarily attributable to the SAVOR-TIMI 53 findings and the analyses were limited by inconsistent outcome adjudication across trials.

GLUCAGON-LIKE PEPTIDE 1 RECEPTOR AGONISTS

Numerous cardiovascular outcomes trials of GLP-1RAs have been completed, with results suggesting within-class heterogeneity of effects. Clear evidence of cardiovascular benefit has been found in the outcomes trials of the GLP-1RA liraglutide, the injected formulation of semaglutide, albiglutide, and dulaglutide (see also Chapters 12 and 13). Although several of the GLP-1RA agents that have undergone cardiovascular outcomes assessments were not found to significantly reduce risk, those agents (including lixisenatide and once-weekly exenatide) did not increase the risk of cardiovascular events compared with aplacebo. Details of the lixisenatide and exenatide once-weekly outcomes trials are summarized in Table 14.2 .

Table 14.2

GLP-1/Dual Incretin Receptor Agonists With Cardiovascular Safety

Reproduced/amended from ADA SOC 2024, Section 10; Adapted from American Diabetes Association Professional Practice Committee. 10. Cardiovascular disease and risk management: standards of medical care in diabetes-2024. Diabetes Care . 2024;47(Suppl 1):S179–S218.

	ELIXA ( n = 6068)	EXSCEL ( n = 14,752)	PIONEER-6 ( n = 3183)	Tirzepatide Meta-Analysis ( n = 7215)
Intervention	Lixisenatide/placebo	Exenatide QW/placebo	Semaglutide oral/placebo	Tirzepatide/active or placebo comparators
Main inclusion criteria	Type 2 diabetes and history of ACS (180 days)	Type 2 diabetes with or without preexisting CVD	Type 2 diabetes and high CV risk (age of ≥50 years with established CVD or CKD, or age of ≥60 years with CV risk factors only)	Varied across trials
A1c inclusion criteria (%)	5.5–11.0	6.5–10.0	None	Varied
Age (years) a	60.3	62	66	58.8
Race (% White)	75.2	75.8	72.3	73.4
Sex (% male)	69.3	62	68.4	56.7
Diabetes duration (years) a	9.3	12	14.9	6.83
Median follow-up (years)	2.1	3.2	1.3	0.85 (all trials at least 26 weeks in duration)
Statin use (%)	93	74	85.2 (all lipid-lowering)	___
Metformin use (%)	66	77	77.4	___
Prior CVD/CHF (%)	100/22	73.1/16.2	84.7/12.2	34.9
Mean baseline A1c (%)	7.7	8.0	8.2	8.29
Mean difference in A1c between groups at end of treatment (%)	−0.3 c	−0.53 c	−0.7	___
Year started/reported	2010/15	2010/17	2017/19	n/a
Primary outcome b	4-point MACE 1.02 (0.89–1.17)	3-point MACE 0.91 (0.83–1.00)	3-point MACE 0.79 (0.57–1.11)	4-point MACE 0.80 (0.57–1.11)
Key secondary outcome b	Expanded MACE (0.90–1.11)	Individual components of MACE (see below)	Expanded MACE or HF hospitalization 0.82 (0.61–1.10)	3-point MACE 0.83 (0.58–1.18); MACE-3 or HF 0.78 (0.56–1.08)
Cardiovascular death b	0.98 (0.78–1.22)	0.88 (0.76–1.02)	0.49 (0.27–0.92)	0.90 (0.50–1.61)
MI b	1.03 (0.87–1.22)	0.97 (0.85–1.10)	1.18 (0.73–1.90)	0.76 (0.45–1.28)
Stroke b	1.12 (0.79–1.58)	0.85 (0.70–1.03)	0.74 (0.35–1.57)	___
HF hospitalization b	0.96 (0.75–1.23)	0.94 (0.78–1.13)	0.86 (0.48–1.55)	___
Unstable angina hospitalization b	1.11 (0.47–2.62)	1.05 (0.94–1.18)	1.56 (0.60–4.01)	0.46 (0.15–1.41)
All-cause mortality b	0.94 (0.78–1.13)	0.86 (0.77–0.97)	0.51 (0.31–0.84)	___

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Insulin Resistance: Pathophysiology, Molecular Mechanisms, and Genetic Insights
2. Effect of Glucose Management on Coronary Heart Disease Risk in Patients With Type 2 Diabetes
3. Cardiovascular and Kidney Effects of Glucagon-Like Peptide 1 Receptor Agonists
4. Pathology and Vascular Biology of Atherosclerosis in Patients With Type 2 Diabetes

Stay updated, free articles. Join our Telegram channel

Join