Summary | Diabetes & CVD: Time for a multifactorial approach
To set the stage, professor Deanfield called diabetes ‘probably the biggest threat to the population’s future cardiovascular health’. It is a sobering fact that a 50-year-old with diabetes, but without history of cardiovascular disease (CVD), has a life expectancy six years less than a counterpart without diabetes [1]. Over the last few years, we have improved at managing the glycemic complications of diabetes, such as hyperglycemic deaths. Cardiovascular (CV) admission rate and the CV complications of diabetes have, however, gone down only modestly [2]. This remains a big challenge.
The risk for our patients with diabetes in terms of CV complications begins before they get frank diabetes. This so-called pre-diabetic phase was nicely shown by the Nurses’ Health Study from the United States, which followed over 100,000 women for 20 years. The results not only showed that if the nurses were diabetic at baseline, their relative risk of myocardial infarction (MI) or stroke was five times higher as compared to if they did not have diabetes. Interestingly, if they suffered a CV event after diabetes diagnosis, their risk was 3.7 times higher, but if they got an event before they had diabetes, their risk was 2.8 times higher. This suggests that the CV impact of dysfunctional glycemic control is already seen before the development of diabetes [3].
This population is increasingly seen in cardiology practice these days. A Swedish study revealed that about one third of patients admitted with an acute MI had diabetes, another third had prediabetes with impaired glucose tolerance, and the remaining third had normal glucose tolerance at the time of their MI [4]. Unsurprisingly, patients with normal glucose tolerance showed better event-free survival (up to 13 years of follow-up). Survival in patients with impaired glucose tolerance, but without diabetes, was decreased, to a similar extent as in patients with diabetes [4].
Another study from Sweden really emphasized the opportunity to manage multiple CV risk factors in these patients and to produce a significant improvement in their clinical outcome. This very large registry-based study in more than a quarter of a million people with type 2 diabetes (T2DM) looked at five potential risk factors [5]. Data showed that if the diabetic persons had all of those risk factors at good level, their risk in terms of MI risk and CV complications was actually not higher than that of the general population. On the other hand, if risk factor control was less optimal, blood pressure (BP), cholesterol, and abnormal glucose control negatively affected risk of acute MI, stroke, and heart failure [5]. Another recent study not only emphasized the risk that originates from having risk factors, but also the opportunity our patients have in terms of improving their CV outcome by adopting a healthier lifestyle. This Chinese study showed that the higher the number of low-risk lifestyle factors, the more protection against future CVD and CV mortality was seen [6].
These insights have induced a change in our approach to diabetes: we no longer only focus on improving blood glucose control. Already in 1971, diabetologists acknowledged that the complications of diabetes are not preventable simply by control of blood sugar, but that much more is needed to get optimal results for patients [7]. Rather than seeing diabetes as a condition that causes CVD, we should consider diabetes as a state of enhanced CV risk. Thus, by modifying the CV risk factors we can achieve substantial benefits for our patients. Management therefore needs to be directed at reducing and delaying CV complications in T2DM patients with and without CVD and in those with prediabetes. The impact of modifying cholesterol and BP in our patients is substantial. For instance, the CARDS Study reported that atorvastatin 10mg lowered the cumulative hazard of MI and CV death by almost 40% within five years in T2DM patients, as compared with control glucose-lowering medication [8].
Despite the fact that statins work so well, many patients remain at substantial residual CV risk, as illustrated by the large Heart Protection Study, in which simvastatin yielded a reduction of major vascular events. Interestingly, patients with diabetes randomized to active treatment had a higher risk than those who did not have diabetes and who received placebo [9]. This suggests a substantial residual CV risk in those with diabetes.
When comparing the benefit of treating patients with diabetes to lower their BP, cholesterol, and blood sugar, it becomes clear that the impact of 0.9% lowering in glycated hemoglobin (HbA1c) on CV events (-2.9% during 5 years of treatment) is smaller than that of lowering LDL-c by 1 mmol/L (-8.2% in 5 years) and lowering systolic BP by 4 mmHg (-12.5% in 5 years)[10].
Additionally, some trials that used active and very intensive glucose lowering therapy in patients, suggested that actually this might be doing harm in terms of CV risk. Notably, the study that investigated the effect of rosiglitazone on the risk of MI, found an increased MI risk when using rosiglitazone in diabetes [11]. As a consequence, the American Food and Drug Administration (FDA) and the European Medicines Agency (EMA) released stringent new guidelines for new diabetes drugs, requiring that any new diabetes drug should demonstrate CV safety in a formal CV outcomes trial. They have to meet a non-inferiority criterion based on comparison with traditional medication.
Thus, since then, a large number of trials have been released that look at the CV impact of new strategies to lower blood glucose. Some of them can be considered real game-changers, in terms of the way we think about CV risk and how we manage it in patients with diabetes. We have got new drugs for lowering blood sugar: the DPP-4 inhibitors and the GLP1 receptor agonists (GLP-1RAs), and also new drugs that affect glucose reabsorption from the kidney, the SGLT2 inhibitors.
The first trial that really surprised the field was the EMPA-REG OUTCOMES Trial, which evaluated the effect of the SGLT2 inhibitor empagliflozin on CV outcomes [12]. Empagliflozin was associated with a substantial clinical benefit; not only was a reduction seen in the primary outcome (14%), but there was also benefit from death from CV causes (38%), death from any cause (32%), and very impressively a rapid benefit in terms of hospitalization from heart failure (35%) [12]. GLP-1RAs came out shortly after that. The LEADER Trial evaluated the GLP1-RA liraglutide on CV outcomes [13]. This agent also met its primary outcome requirements (15% reduction), but in addition showed an improvement in death from any cause (13%).
The pattern of benefit differed between the SGLT2 and GLP-1RA trials, because the improvement with SGLT2 inhibition occurred in the first few months and year, while that of GLP-1RA became clear over the first three and four years of therapy [13]. Thus, these two classes of drugs appear to be working to improve CV outcomes in a different way. Studies are ongoing to understand their mechanisms of action. For now, GLP-1RAs are thought to have anti-atherogenic effects, and SGLT2 inhibitors may mostly exert their effects through hemodynamic effects. Initial observations on combined use of these two classes suggest that the combination yields benefit to lower HbA1c and glucose control [14], but their combined impact on CV complications remains to be assessed.
These two new drug classes have already had a big impact on guideline recommendations on management of T2DM. The new ADA 2018 guidelines, for instance, recommend that patients with diabetes and established atherosclerotic CVD are treated with agents that have been proven to reduce major adverse CV effects and to reduce CV mortality.
However, many questions remain, for example whether these drugs are equally effective in patients without diabetes, or even in patients who have not yet got clinical CVD. Which patients will benefit most from each drug and how shall we be combining these drugs together in clinical practice? This will depend on insights gained in mechanistic research on these drug classes.
Ongoing trials are assessing the effect of these agents in various patient groups. For instance, recent data report on use of the GLP-1RA semaglutide in obese persons without diabetes, showing impressive results in terms of weight loss with semaglutide over a one-year period [15]. This is very exciting concerning opportunities to treat overweight people who are on the trajectory to getting diabetes.
So, we have a new era for CV management in diabetic patients. Diabetologists and cardiologists now have to come together to collaborate on the care of these patients. We have the opportunity to help the patients, not just with BP and cholesterol lowering, but also with these two new classes of drugs which are glucose-lowering drugs, with not yet fully understood CV benefits.
References
Educational information
This is a summary of the presentation given by prof. Deanfield, during the PACE symposium entitled 'Outcomes of GLP-1 RA in Diabetes & CVD: What are the key opportunities for cardiology practice?', held during ESC in Munich, Germany, on August 27, 2018.
Share this page with your colleagues and friends: