Beyond diabetes: a relationship between cardiovascular outcomes and glycaemic index

A relationship between type 2 diabetes and both (i) risk of myocardial infarction (MI) and (ii) worse outcomes after MI is well-established.1 Similarly, the occurrence of type 2 diabetes is associated with high glycaemic index food intake. But less certain has been the relationship between high glycaemic index and adverse cardiovascular events. Such a relationship could have important implications for public health and health policy. Accordingly, Jenkins et al.2 undertook a study of the effects of glycaemic index and glycaemic load on cardiovascular events and mortality in a huge, geographically diverse population of over 137 000 patients spread over five continents, and with a median follow-up of 9.5 years. The study focused on the health effects of foods with a high glycaemic index, a measure of the extent to which 50 g of carbohydrate from that specific food raises the blood glucose level. Several studieshad previously shown that the consumption of a diet with a low glycaemic index can contribute to the prevention and treatment of diabetes. However, the relationship between glycaemic index and cardiovascular risk was less certain. Moreover, most of the prior studies had examined Western populations. Ongoing changes in dietary practices, especially in non-western countries bring a particular urgency to understand the potential impact of these changes on cardiovascular risk. To address these questions, the investigators constructed food frequency questionnaires that were specific to individual countries and even, in the case of India, specific regions. They combined these with detailed lifestyle assessments around educational level, income, employment, smoking and alcohol habits, and cardiovascular health status. The primary outcome was a composite of major cardiovascular events (cardiovascular death, nonfatal MI, stroke, or heart failure or death from any cause) in relation to glycaemic index/glycaemic load. In the primary analysis, participants were also categorized according to the presence or absence of prior cardiovascular disease. The survey showed geographical variations in both glycaemic index (highest in China, followed by Africa and Southeast Asia) and total glycaemic load (highest in South Asia, followed by Africa and China). Data on cardiovascular events were available in almost 120 000 participants with a median time to follow-up of 9.5 years. In comparing the highest vs. lowest quintiles of glycaemic index there was a positive association with the primary composite outcome. In patients with no pre-existing cardiovascular disease, the hazard ratio was 1.21 (95% confidence interval 1.11–1.34) and in those with pre-existing cardiovascular disease hazard ratio was 1.51 (1.25–1.82). The association between the glycaemic index and the primary composite outcome was also significantly stronger among participants with a higher body mass index (BMI) compared with a lower BMI (greater or less than 25). Of course, this is an observational study and, despite the statistical adjustments, it is not possible to draw absolute conclusions on causation, but the emergence of the signal over a study incorporating such enormous cultural and socio-economic diversity is striking. This publication did not report on the associations of glycaemic index/load with other variables of interest, such as total cholesterol, triglycerides, or the incidence of diabetes–but the maximally adjusted statistical models included adjustments for history of diabetes and the use of statins and blood pressure medication. The study also cannot take into account genetic variations, which may also compound susceptibilities to glycaemic index in relation to cardiovascular risk. However, many interesting questions arise. Firstly, there was a stepped increase in hazard ratio for the composite primary outcome across quintiles of BMI in patients with BMI greater than 25. What is the pathobiological mechanism? In other words, how is the glycaemic index translated into a disease process (or processes) that culminate in cardiovascular complications? A number of possibilities could be considered. There is a relationship between local and systemic inflammation and obesity,3 and there may be important considerations in relation to metabolic substrate provision, including to adipose tissue and the nutritional context of that provision.4 Recently, emerging data show that not only does hyperglycaemia drive myelopoiesis and atherosclerosis in experimental animals5,6 but also the pattern of hyperglycaemia is important. Specifically, transient intermittent hyperglycaemia has a particular effect that is transduced by glycolysis-dependent mechanisms in neutrophils, which in turn produce factors that drive myelopoiesis.7 Our own laboratory has shown an effect of hyperglycaemia on epigenetic reprogramming of bone marrow precursor cells, culminating in a trained immunity or immunologic priming state, which persists into macrophages and is pro-atherogenic.8 Clearly the findings of Jenkins et al. have important public health implications. Subsequent trials should establish whether interventions to modify glycaemic index or glycaemic load reduce cardiovascular events in these susceptible populations and there is increasing interest in both caloric restriction and the use of caloric restriction mimetics that seek to manipulate metabolic pathways for homoeostatic gain.9 The study findings add further weight to arguments away from a narrow focus on measures of fasting blood glucose/glycated haemoglobin when considering cardiometabolic risk.8 They also add urgency to the public health advice and policy in respect of food provision, diet, global trends in weight increase, and broader socioeconomic considerations. While these are perhaps mosturgently needed in developing countries, where the rates of change patterns of food consumption may be fastest, Jenkins et al. show that this is a global problem. Conflict of interest: none declared. Biography: Robin Choudhury DM FESC is a Professor of Cardiology at the University of Oxford and a consultant interventional cardiologist at the John Radcliffe Hospital, Oxford. His laboratory works on the interface between metabolic processes and immune function/tissue homoeostasis. He is a Co-PI of the Metabolite-related Inflammation in Diabetes-spectrum Diseases consortium, which brings together investigators from the UK, Sweden, and Denmark. Biography: Naveed Akbar PhD is a senior postdoctoral scientist and clinical lecturer at the University of Oxford. His work focuses on cellular and across-organ communication by extracellular vesicles during sterile injury, inflammation, and during metabolic dysfunction.