Optimising cardiovascular risk management early in the diabetes disease trajectory

PhD in Medical Science, University of Cambridge

Type 2 diabetes increases an individual’s risk of Cardiovascular Disease (CVD). Trials have demonstrated the long term macro-vascular benefits of lowering glucose, as well as other CVD risk factors, in populations with established diabetes. As the diagnostic criterion for diabetes is a threshold on a continuous measure of glucose control, it has been hypothesised that targeting risk factors earlier in the disease trajectory may have even greater e↵ects on rates of complications.

Many nations have introduced programmes to diagnose diabetes earlier, including the NHS in England, where the Health Checks programme o↵ers individuals at ‘high risk’ diabetes testing. This will lead to a greater number of individuals being diagnosed earlier in the course of the disease when treatment decisions are less informed by evidence. Some of the potential harms of intensive treatment are likely to manifest early, while benefits will likely appear years later. Much of the literature relates to lowering CVD risk factors years after diagnosis or arises from studies conducted more than 20 years ago, which may not represent the e↵ects of managing risk factors intensively from diagnosis in contemporary care.

Firstly, I demonstrate that in a population with screen-detected diabetes there is a degree of pharmacotherapy burden at diagnosis, that then intensifies over the following five years.

Figure 3.4. Count of medication types reported in the ADDITION-UK cohort at diagnosis, one and five years. Box-plots represent number of agents, points represent values outside inter-quartile range.
Figure 3.5: Proportion of participants prescribed medication, by agent, in ADDITION-UK from diagnosis to 5 years.

Secondly, I show that there is a large variation in glycaemic control and CVD risk factors after diagnosis at the individual level, which I have characterised and described.

Figure 4.7: Mean (95%CI) HbA1C values at each time point for the four HbA1C trajectory groups identified in ADDITION-Denmark, from diagnosis to five years.
Table 4.2: Comparison baseline characteristics of each identified HbA1C trajectory in ADDITION-Denmark to the preferred low-low and med-low trajectories.
Figure 4.8: Medication use by trajectory group within ADDITION-Denmark (proportions in circles). Prescription medication redemption data is assumed to have 100% coverage. Individuals that died have been excluded from this figure, although patterns of agent redemption are similar including individuals that died during following up (see Figure C.1, on page 226).

Thirdly, I have shown that promoting intensive treatment from the day of diagnosis leads to improvements in cardio-metabolic health, and that increases in medication can decrease the modelled risk of a CVD event.

Figure 5.3: Change in 10-year modelled UKPDS CVD risk from diagnosis to five years, by decile of baseline risk, in ADDITION-Europe. Grey lines represent an individual, and blue lines represent the median and inter-quartile range.
Figure 5.5: Adjusted change in CVD risk factors from baseline to 5 years in ADDITIONEurope, stratified by UKPDS V3 modelled CVD risk. PP=predicted proportion. See Section 5.2.1 on page 75 for details on how estimates within quartiles were conditioned to the entire sample.

Lastly, I show that the risk reduction associated with intensification of pharmacotherapy is not achieved at the expense of quality of life.

Figure 8.2: Distribution of change in HRQoL (EQ-5D, ADDQoL MCS, and PCS) from 0-1 and 1-5 years, colour coded by change in medication.

In conclusion, type 2 diabetes is a disease intertwined with cardio-metabolic health, and actively approaching diabetes care as a multifactorial intervention to improve a cluster of cardio-metabolic risk factors via encouragement of lifestyle change supported by pharmacotherapy is likely to improve health. However, guidelines do not represent diabetes care at the individual level, and further research that both improves our understanding of individual variation and how to communicate these complex relationships will benefit our attempts to further personalise medicine.

comments powered by Disqus