The prevalence of diabetes is taking on epidemic proportions worldwide, largely due to the incidence of insulin resistance associated with increasing levels of obesity and decreased physical activity.1 The increased mortality in diabetes is mainly related to cardiovascular (CV) complications secondary to clustering of risk factors including endothelial dysfunction, dyslipidemia, low-grade inflammation, hypercoagulability, and hypertension.2 Multifactorial intervention reduces CV events in patients at risk, and this approach is currently undertaken in patients with diabetes.3,4 Antiplatelet agents have been recommended to control the hypercoagulable state, but this treatment is not without side effects. In particular, there is an increased risk of gastrointestinal and intracranial bleeding with antiplatelet use, which can have devastating consequences.5 The risk for intracranial bleeding increases in the presence of hypertension, which is commonly found in patients with diabetes. The current review analyzes antiplatelet therapy in diabetes, with special emphasis on the use of these agents in hypertensive patients.
Cardiovascular disease (CVD) remains the main cause of mortality and morbidity in patients with diabetes. Prevention of CVD in diabetes involves a multifactorial approach that aims to treat the cluster of risk factors including hyperglycemia, dyslipidemia, obesity, hypertension, and hypercoagulation associated with this condition. Antiplatelets reduce the prothrombotic environment in diabetes, but complications of this therapeutic approach include a general risk of bleeding, specifically intracranial hemorrhage, the risk of which increases in the presence of hypertension. Current guidelines recommend the use of antiplatelet agents after tight control of blood pressure, which, in clinical practice, is not always possible. In this review, the evidence for antiplatelet use in diabetes with particular emphasis on patients with associated hypertension is examined. Safe levels of blood pressure with antiplatelet therapy, various studies, and general recommendations for diabetes patients, in light of current evidence, are explored. J Clin Hypertens (Greenwich). 2011;13:305–313. © 2011 Wiley Periodicals, Inc.
CV Complications in Diabetes
CV complications in diabetes include coronary artery, cerebrovascular and peripheral vascular disease. The involvement of more than one vascular bed further increases the risk of future events, suggesting that these patients should be treated more aggressively.6 Although the American Diabetes Association indicated more than a decade ago that diabetes should be treated as a coronary artery disease (CAD) equivalent, it was not until relatively recently that the American Heart Association (AHA) adopted this view.7,8 In addition to increasing the risk of first event, the prognosis following cardiac ischemic events in diabetes remains poor despite advances in treatment. Comparing data from 2003 with 1995, work from our division has shown mortality reduction in the nondiabetic population after myocardial infarction, confirming the effectiveness of modern treatment regimens. In diabetes patients, however, mortality was higher and there was little evidence of improvements in outcome, indicating reduced efficacy of current therapies in these individuals.9 Furthermore, even using more advanced treatment strategies, diabetes patients continue to have a significantly worse prognosis. Analyzing patients who underwent angioplasty as a treatment for myocardial infarction has shown that 1-year mortality following an event was 6.8% for those without diabetes and 15.8% for those with diabetes.10 A similar scenario is evident in individuals who undergo coronary artery bypass surgery, as a treatment for extensive CAD, with 1-year mortality rates of 5.9%, 9.7%, and 8.6% in nondiabetic, insulin, and non-insulin–dependent diabetic groups, respectively.11 Given these facts, it appears that diabetes is associated with increased risk of CV disease (CVD) and worse prognosis following an event regardless of the type of intervention undertaken. The increased mortality in diabetes following ischemic events is related to a combination of factors, including more extensive arterial disease and a prothrombotic environment with reduced response to antiplatelet therapy.
Hypertension and CV Complications in Diabetes
Around three quarters of diabetes patients have high blood pressure (BP) and one fifth of hypertensive individuals have diabetes.12 Various mechanisms operate to increase BP in patients with diabetes, including endothelial dysfunction, low-grade inflammation, and abnormalities of the renin angiotensin system.13 Studies have shown that lowering BP in patients with diabetes reduces the risk of future CV events, although the ideal BP target in this population remains under debate. Current guidelines recommend a BP <130/80 mm Hg, particularly in the presence of complications.14 The Action in Diabetes and Vascular Disease: Preterax and Diamicron Controlled Evaluation (ADVANCE) and the Action to Control Cardiovascular Risk in Diabetes (ACCORD) represent two recent major trials in diabetes that investigated the role of BP control on reducing CV events.15,16 The ADVANCE study randomized more than 11,000 individuals to a fixed combination of perindopril and indapamide or matching placebo, and to intensive glucose or standard guideline-based glucose control. Mean entry BP was 145/81 mm Hg, reduced in the active arm by 5.6/2.2 mm Hg compared with placebo (end of follow-up BP was 134.7/74.8 mm Hg and 140.3/77.0 mm Hg, respectively). Reduction in major macrovascular events in the intensive arm compared with standard therapy was 8%, which was not statistically significant, but all-cause and CV mortality were both significantly reduced by 14% (2%–25%) and 18% (2%–32%), respectively. In the ACCORD study, more than 10,000 diabetes patients were randomized to intensive or standard glucose control and were also randomized to a BP-lowering (n=4733) or lipid management intervention (n=5518). Mean entry BP was 139/76 mm Hg, subsequently achieving an impressive systolic BP (SBP) of 119 mm Hg in the intensive arm and 134 mm Hg with standard therapy during 4.7 years of follow-up.17 Despite the major difference in BP, reduction of all CV events was not significant in the intensive arm (12% [−6 to 27%]; P=.20), although major stroke was significantly reduced by 41% (11%–61%) without an effect on coronary events. Taken together, these two studies indicate that BP reduction in diabetes reduces the risk of CV events, although the benefits of very aggressive management are questionable, except for stroke prevention.
A number of studies have attempted to investigate the ideal antihypertensive agents in diabetes, which is beyond the scope of this review and only a quick snapshot is provided. Given the role of the renin angiotensin system in the pathogenesis of CV complications, it is no surprise that studies have shown beneficial effects for angiotensin-converting enzyme (ACE) inhibitors, which are generally regarded as first-line therapy in hypertensive diabetes patients, except for black patients, and with cautious use advised in older patients.18,19 Despite affecting the same pathway, studies investigating the cardioprotective effects of angiotensin receptor blockers (ARBs) have generally been disappointing.20 Nevertheless, these agents are recommended in individuals intolerant to ACE inhibitors. The earlier enthusiasm for combination ACE inhibitor and ARB therapy has waned following the Ongoing Telmisartan Alone and in Combination With Ramipril Global Endpoint Trial (ONTARGET) results. This study included 25,620 patients, 40% of whom had diabetes, and showed that ramipril or telmisartan used separately have similar effects on CV outcome, whereas combination therapy with the two agents does not offer additional benefits despite a further 2.4/1.4 mm Hg reduction in BP. Finally, it appears that addition of a calcium channel blocker to an ACE inhibitor may be superior to an ACE inhibitor/hydrochlorothiazide combination, as the former therapy was associated with 20% reduction in CV outcome in 11,506 patients (60% with diabetes), which may be related, at least in part, to the slightly lower BP by 0.9/1.1 mm Hg achieved by this therapy.21 A number of studies indicate that controlling BP in diabetes decreases the risk of future CV events, regardless of starting BP. However, intensive therapy to reduce BP below 130/80 mm Hg has little effect on combined CV events, although it does reduce the risk of stroke. It should be noted, however, that diabetes is not a single entity but a continuum of different conditions having deranged glucose metabolism as a shared abnormality. For example, CV risk in a diabetes patient with no documented complications is different than that in an individual with proteinuria, and future studies should attempt patient classification to establish the profile of patients who would benefit from intensive BP lowering.
Antiplatelet Treatment in Diabetes and CV Risk
Platelet reactivity is increased in diabetes through a variety of processes that include the effects of insulin resistance and decreased nitric oxide production, glycemia, glycation, and oxidative stress.22 Altered activity of prothrombotic coagulation factors coupled with increased plasminogen activator-1 and impaired fibrinolysis constitute other mechanisms that additionally contribute to increased atherothrombosis risk in this condition.23 Although there is a developing debate about the role of antiplatelet agents in primary prevention in diabetes patients, this therapeutic approach is established for both high-risk primary and secondary prevention of CV events in diabetes.
Aspirin irreversibly acetylates platelet cyclo-oxygenase to inhibit its activity resulting in reduced thromboxane A production, a potent platelet aggregator. Another mode of action for aspirin of potential importance is related to its effect on the coagulation system, resulting in the formation of looser fibrin networks that are easier to lyse.24,25 Aspirin has been used for both primary and secondary CV prevention in diabetes. While the evidence supporting its use in secondary prevention is generally convincing, its role in primary prevention has never been proven and was recently questioned given the negative outcome of various studies.26–30 There are plausible mechanisms for the reduced clinical efficacy of aspirin in diabetes, including interaction between glycation and acetylation in platelets and clotting factors, increased platelet turnover, and generation of thromboxane A through platelet-independent mechanisms.22,31 In general, aspirin is currently not recommended for primary CV prevention in diabetes, and until the results of two major trials are published, A Study of Cardiovascular Events In Diabetes (ASCEND) and Aspirin and Simvastatin Combination for Cardiovascular Events Prevention Trial in Diabetes (ACCEPT-D), aspirin use in diabetes should perhaps be limited for secondary prevention. On the other hand, closer data analysis of various study results suggest that aspirin may be effective in primary prevention in some patients yet to be fully characterized. This uncertainty explains the recent recommendations from the AHA, advocating the use of aspirin for primary CV prevention in high-risk diabetes patients, despite the absence of solid evidence.32
Clopidogrel irreversibly blocks the P2Y12 receptor, thereby inhibiting adenosine diphosphate–induced platelet activation. This agent is usually used as monotherapy in patients intolerant to aspirin or as an add-on to aspirin following an ischemic event. Used in secondary prevention, the Clopidogrel vs Aspirin in Patients at Risk of Ischemic Events (CAPRIE) trial, involving more than 19,000 patients (20% with diabetes), demonstrated that 75 mg/d of clopidogrel was more effective than 325 mg/d of aspirin in reducing ischemic outcomes in the entire group, with an annual event incidence of 5.32% and 5.83%, respectively (P=.04).33 Although the event rate was much higher in diabetes patients, the same pattern was observed (15.6% vs 17.7%, respectively; P=.04), with an effect that was particularly pronounced in higher-risk insulin users.34 Dual therapy with aspirin and clopidogrel following an ischemic event has been shown repeatedly to be superior to aspirin alone, and this strategy has been incorporated into clinical guidelines.31,35 In contrast, the Clopidogrel for High Atherothrombotic Risk and Ischemic Stabilization, Management and Avoidance (CHARISMA) study, involving 15,603 patients (42% with diabetes), has shown that dual therapy is not beneficial in high-risk patients or those with established vascular disease with no history of recent ischemic events.36
In summary, there is some evidence, although inconclusive, to suggest that clopidogrel monotherapy is more effective than aspirin in secondary CV protection in patients with diabetes. Combination therapy of clopidogrel and aspirin is recommended in diabetes patients following an ischemic event. Despite the clinical benefits of clopidogrel, there is evidence to suggest reduced efficacy in diabetes both clinically and biochemically.37,38 In particular, patients with poor metabolic control and impaired renal function and those requiring insulin have the worst response to clopidogrel when used in combination therapy.39–41
The mode of action of this agent is similar to clopidogrel, but metabolism from the inactive to active metabolite is quicker, providing a theoretic advantage. In the Trials to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel–Thrombolysis in Myocardial Infarction (TRITON TIMI 38), which included 13,608 patients, the combination of prasugrel and aspirin was more effective than clopidogrel and aspirin in reducing further events following cardiac ischemia, but the rate of bleeding was higher, which negated the mortality benefit.42 However, when the diabetes group was analyzed separately, the primary end point (composite of CV death and nonfatal CV/cerebrovascular event) was reduced by 30% in the prasugrel vs clopidogrel group (12.2% vs 17.0%, P<.001). Although overall bleeding was higher in the diabetes group, there was no increase in bleeding in prasugrel users.43 One criticism of the study was related to the use of a 60-mg loading dose of prasugrel vs 300-mg clopidogrel, and some argued that results would have been different with a 600-mg loading dose of clopidogrel. However, the Third Optimizing Anti-Platelet Therapy in Diabetes Mellitus (OPTIMUS-3) trial has shown superior platelet inhibition of prasugrel 60/10 mg (loading/maintenance) compared with clopidogrel 600/150 mg in diabetes patients taking chronic aspirin therapy.44 Moreover, higher-dose clopidogrel in combination therapy following an acute coronary syndrome was studied in 25,086 patients in the Clopidgrel Optimal Loading Dose Usage to Reduce Recurrent Events/Optimal Antiplatelet Strategy for Interventions (CURREnT/OASIS7) without showing a benefit in the whole group or in patients with diabetes.45
Therefore, it appears that more effective platelet inhibition with prasugrel beneficial in diabetes and does not increase the risk of bleeding. It can be argued that platelet activation is more pronounced in diabetes and, therefore, more effective platelet suppression would result in better protection without an increased risk of bleeding. The counterargument, however, is related to the fact that diabetes patients are generally at higher risk for bleeding,43 but it appears that more effective suppression of platelet reactivity does not further increase this risk.
This agent is a reversible inhibitor of the P2Y12 receptor, and the recent Platelet Inhibition and Patient Outcome (PLATO) study, involving 18,624 patients, has shown a superior benefit for this agent over clopidogrel when used in combination treatment with aspirin following an ischemic event, without an increase in bleeding rates.46 When diabetes patients (n=4662) were analyzed separately, a similar pattern emerged, with an 11% reduction in primary end point in ticagrelor users, although this failed to reach statistical significance.47 Interestingly, analysis of patients with poor diabetes control demonstrated an enhanced protective effect of ticagrelor, consistent with the reduced efficacy of clopidogrel in the presence of high glucose levels. It should be pointed out that diabetes subgroup analysis in PLATO is underpowered to draw definitive conclusions, but the pattern of results is highly suggestive of a beneficial effect of ticagrelor in diabetes patients.
This agent inhibits phosphodiesterase-3 thereby increasing cyclic adenosine triphosphate concentrations and consequently inhibiting platelet aggregation. Cilostazol was used initially to improve claudication in patients with peripheral vascular disease.48 Recent work, however, suggests a place for this agent in the prevention of ischemic cerebrovascular events, with similar efficacy to aspirin but significantly lower bleeding risk.49 Furthermore, triple therapy with aspirin, thienopyridine, and cilostazol has shown encouraging recent results with reduction in coronary artery restenosis following intervention, particularly in high-risk diabetes patients.50,51 The mechanism for the observed protective effect of cilostazol in diabetes may be related to enhanced inhibition of P2Y12 signaling.52 However, future work and larger outcome studies are needed to confirm these findings before cilostazol is widely used in CAD. Antiplatelet agents and their role in CVD prevention are summarized in the Table.
|Agent||Mode of Action||Primary Prevention||Secondary Prevention||Secondary Prevention After an Acute Event||References|
|Aspirin||Inhibition of cyclooxygenase-1 and modulation of clotting factors||Not effective or effects severely reduced||Effective||Effective in combination therapy||26–32|
|Clopidogrel||Irreversible blockade of P2Y12||Unclear||Possibly more effective than aspirin||Effective in combination therapy with aspirin||33–35,39–41|
|Prasugrel||Irreversible blockade of P2Y12||Not used as monotherapy||Not used as monotherapy||More effective than clopidogrel in combination with aspirin||42–45|
|Ticagrelol||Reversible blockade of P2Y12||Not used as monotherapy||Not used as monotherapy||Probably more effective than clopidogrel in combination with aspirin||46,47|
|Ciloztasol||Inhibitor of phosphodiesterase-3||Not used as monotherapy||Not used as monotherapy||Reduction in target vessel restenosis used in triple therapy||49–51|
Antiplatelet Treatment in Hypertensive Patients With Diabetes: Where is the Evidence?
The main risk factor for primary intracerebral bleeding in all age groups is hypertension,53 followed by amyloid angiopathy in the elderly.54 Nonmodifiable risk factors for intracranial hemorrhage include male sex, older age, and African or Asian origin,55,56 whereas modifiable risk factors include hypertension, smoking, waist to hip ratio, diet, and alcohol intake.57 Of note, diabetes seems to be a risk factor for ischemic but not hemorrhagic stroke. While antiplatelets reduce the risk of atherothrombosis, their benefit is offset by increased risk of bleeding, which is mainly gastrointestinal but can also be intracranial, with the latter resulting in significant morbidity and mortality. To further complicate matters, hypertension itself is associated with increased risk of cerebral ischemic events, and withholding antiplatelet treatment in individuals with suboptimal BP control may have detrimental consequences.
The absolute risk increase of intracranial hemorrhage (ICH) with aspirin is around 1 event per 1000 patients treated,58 and this risk increases in the elderly, particularly in the presence of untreated hypertension.59 Although the risk of ICH with other antiplatelet agents used as monotherapy is less clear, combination therapy with aspirin further increases the risk of bleeding.60 In addition to ICH, a recent study has shown increased subarachnoid hemorrhage (SAH) with new, but not long-term, low-dose aspirin use (odds ratio, 2.52; confidence interval, 1.37–4.62).61 However, since severe cases of SAH with poor prognosis were not included, this study may have underestimated the true prevalence of antiplatelet-induced SAH. Given the increased risk of intracranial bleeding in hypertensive patients, current guidelines recommend withholding antiplatelet treatment until BP is controlled, including in patients with diabetes. Although this sounds logical, there is relatively little evidence to support such an approach, as trials on antiplatelet treatment of hypertensive individuals are limited. The Hypertension Optimal Treatment (HOT) study examined the role of BP control and aspirin in primary CV protection in hypertensive patients. The aspirin arm of the study assigned 9399 patients to 75 mg/d of aspirin and 9391 to placebo. Aspirin treatment was associated with a decrease in major CV events compared with placebo (8.9 and 10.5 events/1000 patient-years, respectively; P=.03) but had no effect on stroke or all-cause mortality.62 Although no increase in intracranial hemorrhage was reported, data were not provided to distinguish between ischemic and hemorrhagic strokes. The authors stated that a benefit of aspirin therapy, albeit marginal, was also seen in hypertensive patients, but no cut-off values were provided. When the subset of 1501 diabetes patients were analyzed, aspirin had similar, and even superior, cardiac protective effects compared with the nondiabetic population, but again had no effect on mortality.
In the Thrombosis Prevention Trial, conducted in 5499 men, the benefit of aspirin for primary prevention was evident only in patients with an SBP below 145 mm Hg. The rate of CV events (coronary heart disease and stroke) in those with SBP >145 mm Hg was 20.5 and 17.9 per 1000 person-years in aspirin- and nonaspirin-treated patients, respectively, compared with 7.7 and 12.2 per 1000 person-years, respectively, for those with SBP <130 mm Hg.63 Therefore, the authors concluded that aspirin treatment is only effective in reducing CV events in patients with SBP <145 mm Hg. Unfortunately, no data were provided on diabetes patients and it is not clear whether the same conclusion applies to this group of patients.
The subgroup analysis of patients with hypertension in the CAPRIE trial, comparing the efficacy of aspirin and clopidogrel, showed a nonsignificant trend toward better outcome with clopidogrel, suggesting that this may be a safer option in individuals with hypertension. A Cochrane review of the use of antiplatelets in hypertensive patients suggests that aspirin fails to reduce events when used for primary prevention in individuals with raised BP and no prior CVD, whereas for secondary prevention it is recommended, as the benefit outweighs the risk.64 In real practice, a significant number of diabetes patients are on antiplatelet treatment with above target BP levels, and it is unclear whether antiplatelets should be withheld in these individuals. The safe BP range in diabetes patients taking antiplatelet therapy is currently unknown, and will probably not be identified in the setting of a randomized controlled trial. However, the continued development of electronic databases in diabetes will shed more light on individuals at risk and may assist in identifying a safe cut-off for BP, above which antiplatelet therapy should not be given.
Recommendations: An Evidence-Based Approach
Aspirin treatment has a role in secondary prevention in CV prevention in patients with diabetes and hypertension. However, the role in primary prevention is debatable, and despite some guidelines suggesting treatment of higher-risk diabetes patients, concrete evidence for such an approach is lacking. It is perhaps advisable to withhold antiplatelet treatment in individuals with an SBP above 145 mm Hg until this is brought under control, although the evidence supporting this is at best scanty. Limited data suggest superior efficacy of clopidogrel monotherapy compared with aspirin in diabetic patients, but this is by no mean conclusive. The role of prasugrel monotherapy in diabetes is unknown but combination treatment with aspirin following an ischemic event is superior to clopidogrel and aspirin without increased risk of bleeding, and the benefit seems to be independent of BP. It should be noted, however, that BP following cardiac ischemia is usually tightly controlled and therefore the use of prasugrel in uncontrolled hypertensive diabetes patients cannot be recommended. The newer antiplatelet agent ticagrelor probably offers an advantage over clopidogrel when used in combination therapy following ischemic events, but, again, evidence is lacking for its use in uncontrolled hypertensive diabetes or as monotherapy. Recommendations for the use of antiplatelet therapy in diabetes are summarized in the Figure.
Despite advances in treatment, the incidence of CVD in patients with diabetes is higher than the general population and the clinical outcome following an ischemic event remains unacceptably poor. Although antiplatelet agents, mainly aspirin, are widely used in diabetes, their safety and efficacy in hypertensive patients is not entirely clear. Aspirin use in patients with diabetes without clinical CVD cannot be recommended, although exceptions may be made in high-risk individuals, with the acknowledgement that the evidence for this approach is far from conclusive. Clopidogrel monotherapy for diabetes patients may be a safer and more effective treatment option, but further work is needed to explore this concept. Aggressive antiplatelet therapy is recommended for diabetes patients following an ischemic event, and the combination of aspirin and prasugrel seems to be particularly effective. BP control is strongly recommended before starting antiplatelet therapy, but safe cut-off values are unclear, and these are left to the discretion of the treating clinician. Practically, it is difficult, and unethical, to conduct antiplatelet therapy studies in diabetes patients with untreated hypertension. However, future observational prospective and retrospective studies may be able to answer the benefit/risk of antiplatelet treatment in diabetic patients with variable BP control.
Acknowledgment: The authors wish to thank the National Institute for Health Research and British Heart Foundation for their continued support.