Abstract
Diabetes mellitus (DM) is also a cause of cardiovascular (CV) disease (CVD). Addressing the atherosclerotic CVD (ASCVD) burden in DM should reduce premature death and improve quality of life. Diabetes mellitus–associated ASCVD can lead to complications in all vascular beds (carotids as well as coronary, lower extremity, and renal arteries). This narrative review considers the diagnosis and pharmacological treatment of noncardiac atherosclerotic vascular disease (mainly in patients with DM). Based on current knowledge and the fact that modern DM treatment guidelines are based on CV outcome trials, it should be noted that patients with noncardiac CVD may not have the same benefits from certain drugs compared with patients who predominantly have cardiac complications. This leads to the conclusion that in the future, consideration should be given to conducting well-designed trials that will answer which pharmacological treatment modalities will be of greatest benefit to patients with noncardiac ASCVD.
Keywords
Introduction
Diabetes mellitus (DM) is a cause of cardiovascular (CV) disease (CVD). Addressing the atherosclerotic CVD (ASCVD) burden in DM should reduce premature death and improve quality of life. 1 Both type 1 and type 2 DM (T2DM) increase the risk of microvascular disease and ASCVD. 2 Diabetes mellitus duration, poor glycemic control, or requiring insulin therapy is associated with vascular disease progression and worse outcomes. 3 The UK Prospective Diabetes Study (UKPDS) reported that the incidence of complications of DM was associated with glycemia. 4 Each 1% fall in glycated hemoglobin (HbA1c) was associated with a 21% reduction in DM-related deaths, a 14% reduction in the rate of myocardial infarction (MI), a 12% reduction in strokes, and a 16% reduction in chronic heart failure (HF). 5 Other risk factors (RFs) are also powerful predictors of CV complications; antidiabetic and antihypertensive drugs affect cardiometabolic variables (eg, body weight, glucose, lipids, kidney function, and serum uric acid levels) differently; these effects should influence selecting drugs for treatment. 6
Atherosclerosis is a progressive disease related to ≥1 RFs and may precede hyperglycemia and the clinical diagnosis of DM. 7 Oxidative stress, glycooxidation, and systemic inflammation, promoted by hyperglycemia, combine to damage arterial endothelial cells and promote arterial inflammation. 7 Atherosclerosis is accelerated in both pre-DM and DM, 3 leading to complications in all vascular beds. 8
The majority of CV outcome trials (CVOTs) highlight the impact of antidiabetic drugs on cardiac events. In this narrative review, we emphasize the impact of DM on noncardiac ASCVD: stroke (with underlying cerebrovascular disease), peripheral arterial disease (PAD), atherosclerotic renal artery stenosis (ARAS), and aortic aneurysms (AAs).
Methods
For this narrative review, we searched English-language publications in PubMed published mostly in the last 2 decades, with emphasis on recent CVOTs involving newer antidiabetic drugs. Our main search terms were diabetes mellitus, vascular risk, pharmacotherapy, stroke, carotid artery disease, peripheral arterial disease, aortic aneurysm, renal artery stenosis. Reference lists of the identified publications were also searched. Clinical trials related to DM treatment with consequent impact on secondary outcomes such as stroke, PAD, ARAS, and AA were included. Also, studies related to the pharmacological treatment of other RFs (eg, hypertension, dyslipidemia, and smoking) and their impact in reducing noncardiac atherosclerotic vascular events are considered.
Noncardiac ASCVD
Stroke
Epidemiology
Major modifiable RFs for stroke include DM, blood pressure (BP), smoking, and dyslipidemia; controlling these RFs can prevent initial strokes as well as their recurrence. 9 Diabetes mellitus is an independent RF for stroke with an incidence 2.5 to 3.5 times higher than in controls; stroke is the most frequent cause of death in patients with T2DM after coronary heart disease. 10 According to a systematic review of 64 cohort studies (775 385 participants; 12 539 strokes), the pooled maximum adjusted relative risk (RR) of stroke associated with DM was lower in men (RR: 1.83, 95% CI: 1.69-2.08) compared to women (RR: 2.28, 95% CI: 1.93-2.69). 11
Diabetes mellitus is associated with a risk of both hemorrhagic stroke (HS) and ischemic stroke (IS). 9 The Emerging Risk Factors Collaboration showed that the adjusted hazard ratios (HRs) with DM were 2.27 (1.95-2.65) for IS, 1.56 (1.19-2.05) for HS, and 1.84 (1.59-2.13) for unclassified stroke. 12 A meta-analysis (39 studies; 359 783 patients) estimated that the prevalence of DM in all patients with stroke was 28% (95% CI: 26%-31%); there was significant heterogeneity (P = .017) in the proportion of DM among the studies; the prevalence of DM was higher in those with IS (33%, 95% CI: 28%-38%) compared with HS (26%, 95% CI: 19%-33%) and in studies that included both stroke types (24%, 95% CI: 20%-28%). 13 A meta-analysis using HbA1c to diagnose DM estimated that the prevalence of DM in all stroke types was 37% (95% CI: 27%-47%). 13 Inadequate glycemic control increases the risk of death from stroke: for every 1% rise in HbA1c, the possibility of stroke-related death is 1.37 higher. 14
Pathophysiology
Stroke in DM is a consequence of the interaction of various factors such as hypertension, systemic inflammation, vascular endothelial dysfunction, increased arterial stiffness, and capillary basal membrane thickening. 9 The underlying systemic inflammation mediates the development of atherosclerosis in DM. 9 Diabetic cardiomyopathy with accompanying abnormalities of early left ventricular diastolic filling may influence the development of HF and rhythm disorders, such as atrial fibrillation, which are stroke predictors. 9 Hyperglycemia in those with and without DM on admission to hospital because of stroke is associated with worse mortality, stroke recurrence, and poststroke outcomes. 15
Types of stroke
The subtype of IS differs between patients with and without DM according to the type of angiopathy induced by DM. 16 Diabetes mellitus was associated with lower relative prevalence of intracerebral hemorrhage (ICH; odds ratio [OR]: 0.63, 95% CI: 0.45-0.9; P = .022), higher relative prevalence of subcortical infarction (OR: 1.34, 95% CI: 1.11-1.62; P = .009), and higher relative frequency of small-vessel (OR: 1.78, 95% CI: 1.31-3.82; P = .012) and large-artery (OR: 2.02, 95% CI: 1.31-2.02; P = .002) disease. 16 In the ICH group, deep hemorrhages were more frequent in patients with DM (66% vs 41%), whereas lobar hemorrhages were more frequent in those without DM (47% vs 26%). 16
Treatment
Until now, no major clinical trials have specifically evaluated stroke prevention strategies in patients with DM, and evidence is poor for secondary stroke prevention. 9
The Action for Health in Diabetes (Look AHEAD) trial evaluated the long-term effects of intensive lifestyle intervention (ILI) in 5145 overweight volunteers with T2DM, and while weight loss and fitness were different between the 2 groups in Look AHEAD, there was no difference in CV end points between ILI and Diabetes Support and Education (DSE; median follow-up: 9.6 years), including stroke as part of the primary composite outcome. 17 The event rate was 1.83 of 100 patient-years for ILI and 1.92 events for DSE (HR: 0.95, 95% CI: 0.83-1.09, P = .51). 17 In the UKPDS study, among patients allocated to intensive blood glucose control, metformin use was associated with a greater effect than sulphonylureas (SU) or insulin for DM-related end points (P = .0034), all-cause mortality (P = .021), and stroke (P = .032). 18 Another meta-analysis (17 trials; 27 705 patients) reported that patients with T2DM on SU treatment had a higher RR for stroke morbidity than those on other drugs (OR: 1.39; 95% CI: 1.16-1.65). 19
Pioglitazone was investigated in patients with T2DM for primary prevention of stroke; compared with patients who did not receive pioglitazone, those on pioglitazone had a lower risk of IS (HR: 0.72; 95% CI: 0.57-0.9), and a significant trend regarding the decrease in IS risk and an increase in pioglitazone dose (P trend = .03) was observed. 20 In a subanalysis of the Prospective Pioglitazone Clinical Trial in Macrovascular Events (PROactive), treatment with pioglitazone was associated with a 47% RR reduction in recurrent stroke (HR: 0.53; 95% CI: 0.34-0.85, P = .0085). 21 Also, in a meta-analysis, using pioglitazone in patients with stroke with insulin resistance, pre-DM, or DM was associated with a lower risk of recurrent stroke (HR: 0.68; 95% CI: 0.50-0.92; P = .01). 22
In patients with T2DM and high CV risk in the Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients–Removing Excess Glucose (EMPA-REG OUTCOME) trial, empagliflozin compared with placebo was not associated with a difference in the risk of cerebrovascular events; nonfatal disabling stroke occurred in 0.2% of the participants in the empagliflozin and 0.3% in the placebo group (HR: 0.82; 95% CI: 0.30-2.26; P = .70); fatal stroke was similar in the empagliflozin and placebo groups (0.3% vs 0.5%, respectively; HR: 0.72; 95% CI: 0.33-1.55; P = .40) as was the composite of nonfatal disabling stroke or fatal stroke (0.6% vs 0.7%, respectively; HR: 0.81; 95% CI: 0.43-1.50; P = .50). 23 There was no significant difference in the risk of transient ischemic attack (TIA; HR: 0.85; 95% CI: 0.51-1.42; P = .54) or the composite of stroke and TIA (HR: 1.05; 95% CI: 0.82-1.35; P = .87). 23
In the Canagliflozin Cardiovascular Assessment Study (CANVAS) program, though canagliflozin reduced CV events by 14% (HR: 0.86, 95% CI: 0.75-0.97, P < .001), active treatment was associated with a nonsignificant reduction in fatal/nonfatal stroke (HR: 0.87, 95% CI: 0.69-1.09, P = .54). 24 The small number of events did not allow to define the effects of canagliflozin on stroke. 25 Regarding stroke subtypes, there was no effect on IS (HR: 0.95; 95% CI: 0.74-1.22), a significant reduction for HS (HR: 0.43; 95% CI: 0.20-0.89), and no effect on undetermined stroke (HR: 1.04; 95% CI: 0.48-2.22). 25
In the Dapagliflozin Effect on Cardiovascular Events (DECLARE) TIMI 58 trial, dapagliflozin compared with placebo was not associated with a difference in the risk of cerebrovascular events; IS was reported in 2.7% in the dapagliflozin and 2.7% in the placebo group (HR: 1.01; 95% CI: 0.84-1.21). 26
The pooled analysis of the 3 sodium glucose transport protein 2 inhibitors (SGLT2is) CVOTs reported a neutral effect on stroke (OR: 1.05, 95% CI: 0.93-1.20, P = .34) and the combined end points of MI and stroke (OR: 0.96, 95% CI: 0.89-1.05, P = .37). 27
A meta-analysis of randomized controlled trials (RCTs) showed that SGLT2i alone compared with placebo did not significantly influence stroke incidence (RR: 1.01, 95% CI: 0.93-1.10, P = .972); no significant differences among 3 SGLT2is were detected: canagliflozin (RR: 0.91, 95% CI: 0.62-1.33, P = .645), dapagliflozin (RR: 0.99, 95% CI: 0.91-1.09, P = .897), and empagliflozin (RR: 1.03, 95% CI: 0.71-1.48, P = .411). 28 A second meta-analysis, which assessed CV outcomes and safety of SGLT2is, disappointingly showed that nonfatal stroke incidence was borderline significantly higher (RR: 1.30; 95% CI: 1.00-1.68; P = .049) in SGLT2i-treated patients compared with placebo. 29 According to the authors, their meta-analysis was limited by the small numbers of events available, particularly for uncommon safety outcomes. 29 There is no definitive evidence that lowering glucose or weight loss increases stroke risk, although it has been suggested that hemoconcentration caused by inhibition of SGLT2 may increase the risk of arterial thrombosis. 29
The pooled data of CVOTs with dipeptidyl peptidase-4 inhibitors (DPP-4i) resulted in a neutral effect on stroke (OR: 0.99, 95% CI: 0.87-1.10, P = .83). 27 Also, pooled analysis of the small RCTs showed a nonsignificant trend toward benefit with DPP-4i regarding stroke (OR: 0.63, 95% CI: 0.33-1.21; P = .170). 30 In contrast, in the analysis of RCTs reporting CV safety, there was no difference in the risk of stroke with DPP-4i treatment compared with placebo (OR: 0.99, 95% CI: 0.85-1.16; P = .958). 30
In the Semaglutide Unabated Sustainability in Treatment of Type 2 Diabetes (SUSTAIN) 6 trial (semaglutide 0.5 or 1.0 mg subcutaneously [SC], once weekly [OW] compared with placebo), nonfatal stroke occurred in 1.6% in the semaglutide group and 2.7% in the placebo group (HR: 0.61, 95% CI: 0.38-0.99; P = .04). 31 In the PIONEER 6 trial (n = 3183 patients, CVOT with oral semaglutide once daily: 14 mg target dose compared with placebo, follow-up: 15.9 months) involving patients with T2DM, first events of nonfatal stroke occurred in 0.8% and 1.0%, respectively (HR: 0.74, 95% CI: 0.35-1.57). 32
Dulaglutide treatment (1.5 mg SC, OW) was associated with a 24% RR reduction in nonfatal stroke (HR: 0.76; 95% CI: 0.61-0.95, P < .015) compared with placebo, in the REWIND (Dulaglutide and Cardiovascular Outcomes in Type 2 Diabetes: a double-blind, randomized placebo-controlled) trial. 33 In the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results (LEADER) trial, the frequencies of nonfatal stroke were lower in the liraglutide group (1.8 mg SC) compared with placebo (3.4% vs 3.8%, respectively; HR: 0.89, 95% CI: 0.72-1.11, P = .30) and fatal stroke (0.3% vs 0.5%, respectively; HR: 0.64, 95% CI: 0.34-1.19, P = .16), although the differences were not significant. 34 In the Exenatide Study of Cardiovascular Event Lowering (EXSCEL) trial, exenatide (dose 2 mg SC) OW did not show improvement in cerebrovascular composite outcome (fatal and nonfatal stroke), compared with placebo (2.5% vs 2.9%, respectively; HR 0.85, 85% CI: 0.70-1.03). 35 Albiglutide treatment OW SC (starting dose 30 mg, up titrated to 50 mg), in the HARMONY OUTCOME (Albiglutide and Cardiovascular Outcomes in Patients With Type 2 Diabetes and Cardiovascular Disease) trial, did not show improvement in fatal or nonfatal stroke compared with placebo (2.0% vs 2.0%, respectively; HR: 0.86, 95% CI: 0.66-1.14, P = .300). 36 However, the Food and Drug Administration has added the risk of anaphylactic reaction to the medication’s label; despite the positive effects of treatment, albiglutide was withdrawn from the market by the manufacturers due to limited prescribing. 37
A pooled analysis showed that treatment with glucagon-like peptide-1 receptor agonists (GLP-1 RAs) significantly reduced the risk of total stroke by 13% compared with placebo (RR: 0.87, 95% CI: 0.78-0.98, P = .021); there was no significant heterogeneity between trials (P = .393, I 2 = 2.30%). 38 Similarly, GLP-1 RA decreased the risk of nonfatal stroke by 12% compared with placebo (RR: 0.88, 95% CI: 0.78-0.99, P = .035); when only fatal stroke was included, active treatment was associated with a nonsignificant reduction by 16% (RR: 0.84, 95% CI: 0.60-1.17, P = .29) and no significant heterogeneity (P = .548, I 2 = 0). 38
Blood pressure treatment is important component of cerebrovascular event prevention. Compared with standard BP control (target systolic BP [SBP] <140 mm Hg), intensive control (target SBP <120 mm Hg) did not reduce total major ASCVD events but did reduce the risk of all strokes (HR: 0.59; 95% CI: 0.39-0.89; P = .01) and nonfatal stroke (HR: 0.63, 95% CI: 0.41-0.96, P = .03), in the Action to Control Cardiovascular Risk in Diabetes Blood Pressure (ACCORD BP) trial. 39 In a stroke outcome meta-analysis of RCTs, intensive BP control was associated with a 17% reduction in the odds compared with the standard control group, with a greater magnitude of benefit in trials in which the SBP was ≤130 mm Hg (P for interaction = .005), in which there was a 47% decrease in the odds of stroke compared with the control group. 40 According the recent European Society of Cardiology (ESC) and the European Association for the Study of Diabetes (EASD) guidelines, in patients with DM, optimal BP control reduces the risk of micro- and macrovascular complications. 41 The BP goal is to target SBP/diastolic BP to <130/80 mm Hg in patients with DM, but not <120 of 70 mm Hg. 41 Dual therapy is recommended as first line, and evidence strongly supports the use of a renin–angiotensin–aldosterone system (RAAS) blocker (angiotensin-converting enzyme inhibitor [ACEI] or angiotensin receptor blocker [ARB] if patients are intolerant to ACEI) and a calcium channel blocker or a diuretic. 41
In a meta-analysis (90 000 randomized patients) including RCTs in which the patients were randomly assigned to a statin or a control group (placebo or usual care treatment), statins reduced the risk of first clinical stroke versus placebo (OR: 0.79; 95% CI: 0.73-0.85). This was largely accounted for by low-density lipoprotein cholesterol (LDL-C) changes; each 10% fall in LDL-C was estimated to decrease the risk of stroke by 15.6% (95% CI: 6.7%-23.6%). 42 The Atorvastatin versus Simvastatin on Atherosclerosis Progression study (n = 325, patients with familial hypercholesterolemia) evaluated the impact of statins (atorvastatin 80 mg vs simvastatin 40 mg) on carotid intima–media thickness (cIMT). After 2 years of treatment, atorvastatin decreased cIMT (P = .0017), while it increased in the simvastatin group (P = .0005). 43 A secondary analysis of the Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial found that patients with T2DM were at higher risk for recurrent stroke (HR: 1.62, 95% CI: 1.33-1.98, P < .001); also there was no difference in the effect of statins in decreasing these events in patients with or without T2DM. 44 Also, a post hoc analysis of the SPARCL trial suggested that statins may increase the risk of ICH in individuals with previous stroke, 45 but these results were not confirmed by 2 meta-analyses. 46,47 An inverse association between high-density lipoprotein cholesterol level and stroke was also reported. 48
The ESC/EASD guideline for DM, pre-DM, and CVD recommended that statins remain “state-of-the-art therapy” for treating dyslipidemia in patients with DM, with ezetimibe or a PCSK9 (proprotein convertase subtilisin/kexin type 9) inhibitor “on top” of a statin—or alone, if there is intolerance to statins—lower LDL-C in patients with DM, thus improving CV outcomes and CV mortality. 41 Recommended targets of LDL-C in patients with DM are: <2.5 mmol/L in patients at moderate CV risk, <1.8 mmol/L or a fall of at least 50% at high CV risk, and <1.4 mmol/L or reduction of ≥50% at very high CV risk. 41
In the ASCEND (A Study of Cardiovascular Events in Diabetes) trial (n = 5480, double blind, randomized, placebo controlled), in patients with T2DM without evident CVD, the primary efficacy outcome (MI, stroke, TIA, or death from any cause) occurred in 658 (8.5%) patients on aspirin (100 mg once-daily) versus 743 (9.6%) on placebo (rate ratio: 0.88, 95% CI: 0.79-0.97; P = .01). Major bleeding occurred in 4.1% patients on aspirin versus 3.2% on placebo (rate ratio: 1.29, 95% CI: 1.09-1.52; P = .003). 49 Zhang et al in a meta-analysis (6 RCTs, n = 11 085 patients) showed that aspirin use did not significantly reduce the risk of stroke, compared with the control group (3.5 vs 4.0%, respectively; HR: 0.83, 95% CI: 0.63-1.10; P = .20). 50 Because of these results, a recent ESC/EASD guideline recommended that aspirin (75-100 mg/d) may be considered in primary prevention in the absence of clear contraindications in patients with DM at high/very high risk, while in patients at moderate CV risk, aspirin for primary prevention is not recommended. 41 In the CAPRIE (Clopidogrel versus Aspirin in Patients at Risk of Ischaemic Events) trial, for the patient subgroup with stroke, the average event rate/year in the clopidogrel group was 7.15% compared with 7.71% in the aspirin group, a RR reduction of 7.3% (−5.7 to 18.7; P = .26). 51 Also, the overall safety profile of clopidogrel was at least as good as that of 325 mg once-daily aspirin. 51 In a post hoc analysis of the CAPRIE trial in patients with DM, the all event rate/year was 15.6% in patients randomized to clopidogrel and 17.7% in patients who received aspirin, with an absolute RR of 2.1% (P = .042). 52 In the Examining Use of Ticagrelor in Peripheral Artery Disease (EUCLID) trial (n = 13 885, double-blind, event-driven trial), ticagrelor (90 mg twice daily) compared with clopidogrel (75 mg once daily) significantly decreased the rate of IS (1.9% vs 2.4%, HR: 0.78, 95% CI: 0.62-0.98, P = .03). 53
In patients with DM, smoking significantly increased the risk of cerebrovascular events and all-cause mortality, whereas quitting is associated with reduced risks; this evidence supports the recommendation of smoking cessation for these patients. 54
Based on the above evidence, it is clear that certain drugs have advantages over others in the primary and secondary prevention of stroke in patients with DM. Each patient should be treated individually, based on the combination of complications, to select the drug(s) which will benefit the most. A more detailed overview of the link between DM and carotid disease is available elsewhere. 55
Peripheral Arterial Disease
Epidemiology
One-third of patients hospitalized because of PAD have DM. 56 Diabetes mellitus and smoking are the most powerful RFs for developing PAD, with ORs of 2.72 and 1.88, respectively 57 ; other important RFs are dyslipidemia, hypertension, and advanced age. 58 In patients with DM older than 40 years, the estimated prevalence is about 20%, 59 increasing to 29% in patients older than 50 years of age. 60
Pathophysiology
Atherosclerosis in patients with DM involves abnormal function of various vascular wall cell types, enhanced coagulation, inflammatory processes, and inhibition of fibrinolysis. 61 These mechanisms also determine the sensitivity of arterial blood vessels to atherosclerosis as well as plaque instability, which may cause thrombosis. 61 Decreased pain sensitivity and diabetic polyneuropathy may delay the diagnosis of PAD in patients with T2DM. 62
Ischemic chronic rest pain of the forefoot (>2 weeks) with or without ischemic lesions or gangrene associated with occlusive atherosclerotic arterial disease defines critical limb ischemia (CLI). 63
Diagnostic tests
The resting ankle–brachial index (ABI) is the initial diagnostic test for PAD for patients with a history or examination suggestive of PAD; the same guideline does not recommend ABI screening in patients who are not at increased risk of PAD. 64 An intermediate group for ABI screening includes patients without suggestive history or examination findings but at increased risk of PAD. 64 An ABI <0.90 is diagnostic for PAD, 56 with 80% sensitivity and 95% specificity. A second test, which is preferable in patients (eg, those with DM) with noncompressible arterial vessels (ABI > 1.3), is the toe–brachial index. 65 It must also be considered that in 30% of patients whose screening tests are normal, the results may become abnormal after exercise. 66 Possible options for this group of patients are walking for 5 minutes on a treadmill (12% grade, 2.0 miles/h), or until symptoms force the patient to stop, 66 and the 6-minute walk test (how far and how fast can the patient walk in 6 minutes). 67 Furthermore, Doppler ultrasonography (DU) has become established, while other imaging methods such as computed tomography angiography (CTA), magnetic resonance angiography (MRA), and angiography are mainly used for preoperative work-up. 63
Treatment
The goals of medical therapy for PAD include improving limb symptoms, exercise capacity, and quality of life, as well as reducing the risk of adverse CV events and limb events. 68 This treatment for patients with T2DM and PAD is similar to that recommended in guidelines for patients with CVD in general. 41 Given the aforementioned importance of DM as an RF in the development of PAD, treatment of hyperglycemia is a significant component of prevention, although data are scarce in recent CVOTs.
In the meta-analysis of 9 RCTs (n = 10 897, patients with T2DM), intensive (HbA1c 6%-7.5%) versus less intensive control showed significant reduction in the risk of amputation (RR: 0.65; 95% CI: 0.45-0.94; I 2 = 0%). 69 Also, the UKPDS showed that the glycemia-associated reduction in risk for microvascular end points and for amputation or death from PAD was greater (by 37% and 43% per 1% reduction in HbA1c concentration, respectively, each P < .0001) than it was for MI, stroke, and HF (by 14% [P < .0001], 12% [P = .035], and 16% [P = .021] per 1% HbA1c, respectively). 5 In the posttrial monitoring of UKPDS, no significant RR reduction was observed for PAD (P = .19). 70 Also, it was pointed out that the conclusions of the UKPDS study cannot directly be related to patients with DM and PAD, because although many such patients were included, the prevalence of PAD was not clearly defined in that study. 71 An increase in HbA1c level by 1% leads to a 21% increase in risk of PAD during an average follow-up of 4.6 years in the Bypass Angioplasty Revascularization Investigation in T2DM (BARI 2D) trial. 72
Results from the meta-analysis (4 RCTs, n = 990) showed that, in patients with T2DM, metformin treatment decreased the risk of PAD compared with the control group, with limited heterogeneity between studies, but did not achieve significance (HR: 0.81, 95% CI: 0.50-1.31, P = .39). 73 In the Action in Diabetes and Vascular Disease: Preterax and Diamicron Modified Release Controlled Evaluation (ADVANCE) trial, there was no significant difference between intensive (gliclazide-modified release plus other drugs as required to achieve HbA1c ≤6.5%) versus standard treatment for peripheral vascular events (6.2% vs 6.6%, respectively; P > .05). 74 In the PROactive trial, pioglitazone treatment led to a more frequent but nonsignificant risk of leg revascularization, compared with placebo (3.07% vs 2.46%, HR: 1.25, 95% CI: 0.90-1.73). 75
Peripheral arterial disease was not fully investigated in most recent CVOT studies with GLP-1 RA and SGLT2i, in contrast to cerebrovascular and CV outcomes. The perception of the effect of SGLT2i on PAD was changed after the results of the CANVAS program, in which treatment with canagliflozin compared with placebo increased the risk of amputation of toes, feet, or legs (HR: 1.97; 95% CI: 1.41-2.75); the highest rate of amputation was at the level of the toe or metatarsal in 71% of those who underwent an amputation. 24 Treatment with empagliflozin compared with placebo, in the post hoc analysis of the EMPA-REG trial, did not change the risk of lower limb amputation (LLA; HR 1.00, P = .99), with similar results with both doses (10 and 25 mg). 76 Also, a pooled analysis (data from phase IIb/III clinical trials) showed that dapagliflozin did not increase the risk of LLA versus placebo (P > .05). 77 Recent data from RCTs showed that SGLT2i as a class did not increase the risk of amputation significantly (OR: 1.40, 95% CI: 0.81-2.41), but subgroup analysis in patients treated with canagliflozin compared with placebo or other oral antidiabetics showed significant increase in the risk of amputation (OR: 1.89, 95% CI: 1.37-2.60). 78 Based on recent data, an expert panel concluded that there may be an increased amputation risk related only to canagliflozin, representing a specific drug effect rather than a SGLT2i class effect. 79
Based on limited data, treatment with GLP-1 RAs did not negatively affect the risk of PAD. Treatment with liraglutide (the post hoc analysis of LEADER trial, liraglutide SC 1.8 mg) did not increase the risk of a DM-related foot ulcer (HR: 0.92, P = .41) but did decrease the risk of amputations (HR: 0.65, P = .03) versus placebo. 80 Semaglutide treatment (0.5 or 1.0 mg SC OW) reduced the risk of coronary and peripheral revascularization (HR: 0.65, P = .003), versus placebo, without data dedicated to amputations. 31 Also, exenatide treatment (from the post hoc analysis of EXSCEL trial) did not affect the risk of lower limb composite events (nontraumatic amputations, gangrene, and endovascular/surgical revascularization procedures) in both patients with (HR: 0.99) and without PAD (HR: 0.96, for interaction P = .92). 81
There is limited evidence about the impact of DPP-4i on PAD. Chang et al in their retrospective nationwide cohort study (mean age: 58.9 ± 12.0 years; 54% were male) showed beneficial effects in lowering the risk of PAD and limb amputations in patients with T2DM. 82 During a mean 3.0-year follow-up, 3369 DPP-4i users and 3880 DPP-4i nonusers were diagnosed with PAD. Compared with nonusers, DPP-4i users were associated with a lower risk of PAD (HR: 0.84, 95% CI: 0.80-0.88). Additionally, DPP-4i users had a decreased risk of lower extremity amputation compared with nonusers (HR: 0.65, 95% CI: 0.54-0.79). 82
Quitting smoking can prevent both the onset and worsening of PAD. 71 In an observational cohort analysis (after 5 years of follow-up), patients who stopped smoking, compared with those who continued, had a significantly improved all-cause mortality (14% vs 31%; HR: 0.40; 95% CI: 0.18-0.90) and amputation-free survival (81% vs 60%, HR: 0.42, 95% CI: 0.22-0.86). 83
There is compelling trial-based evidence showing that lipid-lowering treatment is beneficial in patients with PAD for both primary and secondary prevention. 84 The beneficial effect of statins might be partially attributed to their pleiotropic effects (eg, anti-inflammatory and antiatherogenic activities) beyond cholesterol lowering. 85
According to recent guidelines, statins are recommended for all patients with PAD. 64 A nationwide population-based study (n = 69 332 patients with DM and PAD; 11 409 statin users; 4430 non statin users, 53 493 lipid-lowering agent nonusers) showed that statin users compared with nonusers had a lower risk of lower extremity amputation (adjusted HR: 0.75; 95% CI: 0.62-0.90). 85 Treatment with statin (simvastatin 40 mg once daily) versus placebo, in the Heart Protection Study (recruited n = 6748 adults with PAD plus n = 13 788 high-risk patients), showed a 16% RR reduction of first peripheral vascular event (4.7% vs 5.5%, respectively, P = .006) among all participants in the study, irrespective of other RFs. 86 In the FOURIER (Further Cardiovascular Outcomes Research with PCSK9 Inhibition in Subjects With Elevated Risk) trial (n = 27 564, randomized placebo-controlled, follow-up median: 2.2 years), treatment with evolocumab versus placebo showed the RR reduction in major adverse limb events (the composite of major amputation or urgent peripheral revascularization for ischemia) in patients with PAD (HR: 0.63; P = .063) and in the overall population (HR: 0.58; P = .0093). 87 In the Fenofibrate Intervention and Event Lowering in Diabetes (FIELD) study (n = 9795, 5-year follow-up, randomized placebo control), treatment with fenofibrate (200 mg once daily) reduced the risk of first amputation (HR: 0.64, P = .02) and minor amputation (HR: 0.53, P = .027), compared with placebo. 88
Target BP is defined by the ESC/EASD guideline as for other CVDs, and the main classes of drugs that are recommended are RAAS blockers. 41 The American College of Cardiology (ACA)/American Heart Association (AHA) guidelines give a class II recommendation for ACEI or ARBs for PAD. 64 Also, in patients with PAD, treatment with β-blockers is not contraindicated and does not worsen claudication symptoms. 89 Nevertheless, the true benefits of different classes of antihypertensives for PAD certainly require well-designed studies.
The choice of antiplatelet therapy in PAD is a source of debate. In the Critical Leg Ischaemia Prevention Study (CLIPS; n = 366, randomized placebo-controlled, double-blind), groups on aspirin (100 mg daily) versus placebo showed significant reduction in critical leg ischemia (P = .014). 90 In contrast, the Prevention of Progression of Arterial Disease and Diabetes (POPADAD) trial (n = 1276, randomized, double-blind, placebo-controlled trial), in patients with DM treatment with aspirin (100 mg daily) versus placebo, did not show benefits in amputation for CLI (HR: 1.23, 95% CI: 0.51-2.97, P = .64). 91 For patients with PAD, clopidogrel compared with aspirin treatment showed an RR of 23.8% in reducing the risk of vascular events, in favor of clopidogrel (P = .0028) in a post hoc analysis of the CAPRIE trial. 51 Single-agent antiplatelet therapy is a 2016 ACA/AHA class Ia recommendation; however, the guidelines do not recommend clopidogrel over aspirin. 64 In patients with symptomatic PAD, in the EUCLID trial, ticagrelor compared with clopidogrel did not decrease the rate of hospitalization for acute limb ischemia (1.7% vs 1.7%, respectively, HR: 1.03, 95% CI: 0.79-1.33, P = .85) and lower limb revascularization (12.2% vs 12.8%, respectively, HR: 0.95, 95% CI: 0.87-1.05). 53 The 2016 ACC/AHA PAD guidelines provide a class IIb recommendation for dual antiplatelet therapy with clopidogrel, suggesting it might be considered for selected patients but that providers should be cautious given the absence of substantial long-term evidence. 64
A number of trials evaluated the influence of cilostazol on maximal walking distance (MWD). A study (216 patients with and 599 without DM, study duration varied from 12 to 24 weeks) investigated cilostazol treatment in patients with DM. 92 They showed that among patients with and without DM, cilostazol (100 mg twice daily) was superior to placebo (estimated treatment effect on MWD: 1.15, 95% CI: 1.05-1.25, P = .001, and 1.24, 1.18-1.31, P < .0001, respectively); there was no significant difference in increased percentage of MWD (51% vs 61%) between the DM and non-DM patients treated with cilostazol. 92 Among patients with more severe claudication (shortest baseline MWD), those with DM experienced a greater % increase in MWD with cilostazol than those without DM (34% change from baseline MWD in Q1 compared with 5.5%, 23%, and 17.2% in Q2, Q3, and Q4; log-transformed data). 92 There was no significant difference in the adverse event profile of the DM and non-DM patients on cilostazol; no excess hemorrhagic events occurred in cilostazol-treated patients with DM. 92
Although PAD is a common manifestation of DM, it is often unrecognized and undiagnosed. Results from trials on the specific effects of antidiabetic drugs on PAD are scarce. Analyses of large clinical studies have demonstrated that using lipid-lowering drugs and antihypertensives reduces the CV risk in patients with PAD and DM, with less clear effects for antiplatelet agents.
Atherosclerotic Renal Artery Stenosis
Epidemiology
Diabetes mellitus leads to diffuse atherosclerotic lesions; therefore, it may be expected that the prevalence of ARAS is increased in patients with DM and hypertension. 93 Based on an autopsy study (n = 5194, macroscopically apparent stenosis of >50% of the vessel lumen was described as ARAS), DM was present in 103 patients in the ARAS group (52%; 95% CI: 45%-59%) and in 62 patients (24%; 95% CI: 19%-29%) in the random sample. 94 Furthermore, the calculated frequency of ARAS in patients with DM was 8.3% (95% CI: 6.8-9.9); the OR being 3.5 (95% CI: 2.6-4.6), and the frequency of ARAS in patients with DM and hypertension reached 10.1% (95% CI: 8.0%-12.20%). 94 Bilateral ARAS was found in 30% of non-DM patients with ARAS and in 43% of patients with ARAS and DM, although this difference was not significant (P = .059); also, there were no significant differences between patients with DM with and without ARAS regarding age, sex, or DM duration. 94 Postma et al reported that ARAS (using imaging of the renal arteries by MRA) was found in 18 (33%) of 54 patients with DM, of which 12 (22.2%) were over 50% and 6 (11.1%) had an ARAS of <50% of the luminal diameter. 93 Benjamin et al reported no significant association between ARAS and DM (adjusted OR: 0.63, 95% CI: 0.34-1.19, P = .158). 95 Also, in other studies, DM was not associated with ARAS or the differences did not reach significance. 96,97 Although DM is a major cause of atherosclerosis, a strong association with ARAS may not exist. 95
Clinical signs and diagnosis
Atherosclerotic renal artery stenosis is associated with renovascular hypertension and nephropathy. 98 When occlusion is >70% to 80% of the luminal area, this is considered as “critical stenosis” because it induces renal hypoperfusion and activates the RAAS. 99
Hypertension, often resistant to treatment, is the main clinical manifestation of ARAS. An acute rise in serum creatinine (>30%) may follow ACEI and ARBs administration due to blocking compensatory mechanisms. 100 Also, recurrent episodes of flash pulmonary edema (Pickering syndrome) may indicate bilateral ARAS. 100
The diagnosis of ARAS is based on imaging and functional tests. The significance of ARAS is traditionally based on estimating anatomical stenosis using renal DU (RDU) or 2-dimensional imaging, including angiography, CTA, and MRA. As an initial test, ultrasound can be used; >1 cm difference in size between the 2 kidneys may indicate ARAS. 100 Peak systolic velocity (PSK >300 cm/s) measured by RDU is the most accurate test (specificity: 92%, sensitivity: 85%, and diagnostic OR: 60.9%). 101 Radermacher et al showed that a renal resistance index (a sonographic parameter defined as PSK − end diastolic velocity/PSK) ≥80 was a strong independent predictor of renal disease progression and could identify patients who may benefit from angioplasty or surgery. 102 Angiography, previously considered as the gold standard for estimating luminal stenosis, is far from optimal; atherosclerotic involves multiple areas of stenosis or poststenotic dilation, which may affect the accuracy of luminal reduction estimation due to the lack of a reference segment. 103 Because of that, both CTA and MRA may overestimate the severity of ARAS. 103 Therefore, BOLD-MRI (using paramagnetic properties of deoxygenated hemoglobin, can assess the ability of renal tissue to extract the oxygen from the blood) may provide useful information to identify the kidneys that could benefit from revascularization. 100
Diabetes mellitus is an RF for contrast-induced acute kidney injury (CI-AKI). 104 This risk may be increased if ARAS is also present. Also, the use of metformin should be considered in patients with DM receiving contrast media (CM). 105 Assessing the risk of CI-AKI before intravascular CM administration is best carried out by measuring estimated glomerular filtration rate (eGFR), but a questionnaire may detect most patients with eGFR <45 mL/min/1.73 m2. 105 Volume expansion with isotonic saline or sodium bicarbonate remains the mainstay of CI-AKI prevention, although there is still uncertainty about the optimal protocol. 105
Treatment
Although recent CVOTs followed renal outcomes, ARAS was not investigated as a separate outcome. If antihypertensive drugs are considered, according to guidelines, ACEIs (class Ia), ARBs (class Ib), calcium channel blockers (class Ia), and β-blockers (class Ia) are effective for the treatment of hypertension associated with unilateral ARAS. 106 It is also necessary to mention that in cases of bilateral ARAS and impaired baseline renal function (serum creatinine <200 μmol/L), the introduction of ACEI treatment is highly likely to switch off most of the GFR; creatinine will rise significantly, usually within a few days. 107 Angiotensin-converting enzyme inhibitor-induced renal impairment is a sensitive indicator of bilateral ARAS. 107 It is also relevant to consider that with advanced ARAS (especially with bilateral involvement), the use of ACEI (or ARB) may permanently damage the kidneys. 108,109 A meta-analysis showed that revascularization was of marginal benefit and just slightly reduced the need for antihypertensive medications in patients with ARAS, hypertension, and/or chronic kidney disease. 110 Revascularization compared with medical therapy did not reduce adverse CV or renal outcomes over an average follow-up of 34 months. 110 Currently, the only class I recommendation for renal revascularization is in the setting of recurrent pulmonary edema. 106
Drummond et al reported that active smoking results in clinically significant ARAS, and adverse cardiorenal events, at much younger age; clearly, there is a need for smoking cessation. 111 Overall, the optimal treatment for ARAS remains poorly defined.
Aortic Aneurysm
Epidemiology and pathophysiology
Aortic aneurysm represents a localized enlargement of the aorta resulting from weakening of the aortic wall; the abdominal infrarenal aorta is the most common site, although AAs can occur at other sites of the aorta (and in other arteries). 112 The thoracic and abdominal aorta differs in structure, biochemical composition, smooth muscle cell origin and biology; although abdominal AA (AAA) and thoracic AA (TAA) share common mechanisms, they also have distinct features. 113 In the proximal part of the aorta (above the ligamentum), an AA is generally nonarteriosclerotic, smooth, noncalcified, and strongly genetically mediated. In contrast, in the distal part (below the ligamentum), it is irregular, calcified, arteriosclerotic, and related to established arteriosclerotic RFs. 114
Thoracic AAs frequently develop at younger ages, are often linked to genetic mutations (eg, Marfan syndrome or Loeys-Dietz syndrome), and are frequently associated with aortic dissection. 115 Altered extracellular matrix and vascular smooth muscle cell (VSMC) tone play important roles in TAA, but the inflammatory component and VSMC apoptosis are less prominent than for AAAs. 113,115,116 Hypertension and atherosclerosis are important RFs for the development of TAA and AAA. 117 Despite the differences between TAAs and AAAs, epidemiological studies have reported paradoxical inverse relationship with DM. 118 The majority of these studies identified an inverse association between DM and the prevalence and incidence of AAA. 119 -121 Although the underlying mechanisms are not established, it has been suggested that DM is associated with excess vascular matrix, which could protect against the loss of arterial wall matrix typically seen in AAA. 122 A meta-analysis of 16 cohort studies (>16 000 cases of AAA among 4.5 million participants) suggests that individuals with a diagnosis of DM have a 42% reduction in risk of AAA. 122 A meta-analysis reported a significant negative association between DM and TAA (OR: 0.70; 95% CI: 0.59-0.84), similar to that of AAA. 123
A meta-analysis (21 cohort studies, n = 5 440 588 participants) suggested that hypertension increases the risk of developing AAA by 66% 124 but is not associated with AAA expansion. 125 Individuals with DM have excess vascular matrix (increased synthesis and reduced degradation of matrix) via advanced glycation end products 119,126 and thicker aortic walls which may reduce wall stress and protect against the development of AAA. 127 However, DM is associated with poorer short- and long-term outcomes after AAA repair. 128 Survival (short or long term) is poorer in patients with DM suggesting an increased CV burden regardless of the treatment administered for AAAs. 128
Treatment
Epidemiological studies have suggested that the protective effect of DM on AAA is not only related to the pathophysiology of DM but also to the treatment for DM. 112 A study including 1269 patients with AAA (94.1% men, mean age: 67 years, follow-up: 3.4 years) showed that using antidiabetic drugs was associated with a 56% reduction in AAA growth rate; this association was independent of confounding factors including other drugs. 129
Metformin usage was negatively associated with AAA enlargement and remained significant after controlling for gender, age, cigarette smoking status, and obesity. 130 Also, metformin use does not change the risk of ruptured AAA in patients with DM. 131 To date, the literature regarding the association between the use of DDP-4 inhibitors and AAA development in humans is poor. Also, no clinical study has as yet reported the association between the use of GLP-1 RA and AAA prevalence, incidence, or progression. A case–control study using a population database (n = 1.2 million patients with T2DM) showed that metformin, thiazolidinediones, and SU had protective effects on AAA development, but not DPP-4i or α-glucosidase inhibitor. 132
A meta-analysis (14 RCTs, n = 38 749) showed that statin treatment was associated with significantly slower AAA growth rate (mean difference = −1.5 mm/year, P < .00001), compared with placebo. 133 Also, statin treatment showed benefits in the short- and long-term mortality after AAA repair (for both P < .00001), compared with placebo or other conventional treatments. 133
The effectiveness of β-blockers in aneurysm disease is controversial, but this treatment has become standard practice. 114 Lindeman and Matsumura suggested that the available clinical studies refute β-blockers or ACEI as pharmaceutical strategies for AAA stabilization, and this indirectly confirms absence of a direct association between BP and AAA progression. 134 On the other hand, ARBs and potassium-sparing diuretics were associated with slower AAA growth rates. 129 In a meta-analysis (n = 15 475, 18 studies), mean BP had no effect and antihypertensive or other cardioprotective medications only had small nonsignificant effects on AAA growth. In contrast, current smoking was associated with an increased rate of expansion (0.35 mm/year), which is twice as fast as growth in previous or nonsmokers. 135
Currently, AAA patients are recommended to receive low-dose aspirin, but compelling evidence is lacking. 118 The specificities of individual noncardiac ASCVD entities in contrast with those of coronary artery disease are shown in Table 1.
The specificities of individual non-cardiac ASCVD entities in contrast with those of coronary artery disease (CAD)
Abbreviations: AA, aortic aneurysm; AAA, abdominal aortic aneurysm; ABI, ankle–brachial index; ACEI, angiotensin-converting-enzyme inhibitors; ARAS, atherosclerotic renal artery stenosis; ARB, angiotensin receptor blocker; ASCVD, atherosclerotic cardiovascular disease; BP, blood pressure; CAD, coronary arterial disease; CI-AKI, contrast-induced acute kidney injury; cIMT, carotid intima–media thickness; CTA, computed tomography angiography; DBP, diastolic blood pressure; DM, diabetes mellitus, DU, Doppler ultrasonography; GLP-1 RA, glucagon-like peptide-1 receptor agonist; HF, heart failure; ICH, intracerebral hemorrhage; MI, myocardial infarction; MRA, magnetic resonance angiography; MWD, maximal walking distance; PAD, peripheral arterial disease; SBP, systolic blood pressure; SGLT2i, sodium-glucose transport protein 2 inhibitor; SU, sulphonylurea; 3-P MACE, 3-point major adverse cardiovascular events; TBI, toe–brachial index.
Conclusions
In recent years, CVOTs have been predominantly concerned with cardiac complications with less attention (and often number of events) to vascular complications elsewhere (eg, stroke, PAD, ARAS, or AA). Commonly, various CV complications are classified as a single entity. However, each noncardiac ASCVD patient is characterized by different RFs, which lead to the development of specific vascular complications. Therefore, each patient is unique.
It is also important to emphasize that in the era of modern diagnostic tools, we must not forget the clinical examination. For example, carotid auscultation is an inexpensive diagnostic tool that could provide useful information. 136 If suspicion is raised for any clinical entity using clinical examination, that possibility should be investigated. 136
Achieving adequate glycoregulation, without losing weight, may decrease the prevention of CV complications compared with treating T2DM together with significant weight loss. 137 There is evidence that weight-centric treatment of T2DM is associated with CVD prevention, but there is also a need to establish if weight loss improves the treatment of noncardiac ASCVD. 138
Based on current knowledge and the fact that modern DM treatment guidelines are based on CVOTs, it should be noted that patients with noncardiac ASCVD may have different benefits from certain drugs compared with patients with predominantly cardiac disease. Consideration should be given to conducting well-designed trials to guide the pharmacological treatment of patients with noncardiac ASCVD.
Footnotes
Author Contributions
Emir Muzurović contributed to conception and design, wrote the manuscript, contributed to the research data discussion, and reviewed the data and manuscript. Dimitri Mikhailidis contributed to conception and design, drafting, and critical review of the manuscript. All authors approved the final version of the text.
Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: E.M. has given talks or attended conferences sponsored by Novo Nordisk, SANOFI, AstraZeneca, SERVIER, and Merck. D.P.M. has given talks, acted as a consultant, or attended conferences sponsored by Amgen, Novo Nordisk, and Libytec.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
