Type 1 diabetes mellitus is a chronic, metabolic auto-immune disease characterized by the destruction of pancreatic β-cells, resulting in absolute insulin deficiency (i.e., little to no insulin production) and leading to difficulties in maintaining blood glucose homeostasis. Due to their unfavourable and challenging glycaemic condition, people with type 1 diabetes face a significant disease burden with a reduced quality of life.

This doctoral thesis focuses on three main concepts related to type 1 diabetes management, which are introduced in Chapter I. The first key concept is glycaemic control as maintaining near-normal blood glucose is of utmost importance to delay the onset of diabetic complications. HbA1_c has long been considered the gold standard to evaluate glycaemic control, however nowadays continuous glucose monitoring (CGM) has become the standard of care for glycaemic management in type 1 diabetes clinical practice, measuring interstitial glucose levels throughout the entire day. Second, our research focuses on cardiovascular (CV) health. Despite advances in type 1 diabetes care, individuals with type 1 diabetes are still at an overall increased CV risk compared to their non-diabetic peers. The number one cause of mortality in type 1 diabetes is cardiovascular disease (CVD) however optimal screening strategies remain unknown. Arterial stiffness might be a valuable proxy for CVD and assist in detecting patients with ‘early vascular aging’ and hence, increased CV risk. Carotid-femoral pulse wave velocity (cf-PWV) is the gold standard to evaluate arterial stiffness and has shown prognostic value for CVD and mortality in the general population. A third concept is physical activity (PA), which can play an important role in type 1 diabetes management as longitudinal and interventional studies have shown that physically active patients show fewer CV risk factors, fewer diabetes-related complications, better well-being, and increased longevity. However, exercise-associated dysglycaemia defers many people with type 1 diabetes from being physically active.

Summarized, the integrated aims of this doctoral thesis were to enhance our understanding of glycaemic control and gain insight into arterial stiffness, in order to respectively improve glycaemic management – in particular around physical activity – and refine CV risk stratification.

The first specific aim was to gain insight into the concept of glycaemic control and examine the added value of emerging CGM-derived parameters to clinical management, and its possible contribution to the prediction of vascular outcomes.

In Chapter II, we explored key CGM-parameters of glycaemic control currently used in the clinical management of type 1 diabetes (i.e., time in range [TIR] and coefficient of variation [CV%]) in almost hundred patients (CARDIA 1.0-study). The added value, interrelation and interpretative value of these CGM-parameters were examined in patients stratified according to their level of glycaemic control by means of HbA1_c. We confirmed the shortcomings of HbA1_c and demonstrated that TIR and CV% are valuable complements to reflect glycaemic control and variability, respectively. TIR and CV% were related to clinically important situations, TIR being mostly dependent on hyperglycaemia and CV% being reflective of hypoglycaemic risk. However, we also warranted a critical approach for their individual interpretation and applicability. TIR can neither be used as the sole measure of glycaemic control as it did not provide information on hypoglycaemia, and the clinical meaningfulness of CV% should be carefully considered. Importantly, the relationship between TIR and CV% seemed to depend on the individual patient, with CV% generally adding less clinically relevant information in those with poor glycaemic control. We pointed to the need for further research and evaluation of composite measures of glycaemic control.

Regarding the possible role of CGM-parameters in the prediction of complications, in Chapter IV we evaluated the relationship of several CGM-metrics (TIR, time above range, time below range, glycaemic variability parameters) with arterial stiffness in a cohort of fifty-four patients (CARDIA 2.0-study). While cf-PWV was associated with HbA1_c, no relationship was found with any of the short-term CGM-parameters. We concluded that HbA1_c should continue to be used in the risk assessment for diabetic complications next to other glycaemic control parameters such as TIR, complementing instead of competing with each other.

The second aim was to explore the concept of arterial stiffness (evaluated by cf-PWV) as a biomarker that could aid CV risk stratification in type 1 diabetes, as well as to investigate determinants of arterial stiffness in type 1 diabetes.

In Chapter III, we assessed cf-PWV in fifty-four patients with long-standing type 1 diabetes and still free from known CVD (CARDIA 2.0-study). Additionally, cf-PWV was compared to other CV risk evaluation tools used in type 1 diabetes. We found that a substantial proportion (24%) of patients showed premature arterial stiffening, which was strongly associated with the STENO risk score for future CV events, thereby illustrating the clinical relevance of arterial stiffness. Importantly, however, considerable heterogeneity was observed in arterial stiffness, STENO score, and hence in CV risk, and differences in risk categorisation between the STENO tool and European Society of Cardiology (ESC) guideline-based criteria. We thereby pointed to the need for refinement of CV risk classification in type 1 diabetes and indicated that studies need to investigate if evaluation of arterial stiffness should be implemented in clinical practice.

In Chapter IV, predictors of arterial stiffness (cf-PWV), including traditional CV risk factors and different time-varying parameters of glycaemic control and glycation, were investigated in the same cohort (CARDIA 2.0-study). In these patients with long disease duration but free from overt CVD, longer-term glycaemic exposure as reflected by current and mean 10-years HbA1_c was a key predictor of arterial stiffness, next to traditional CV risk factors age and blood pressure, as well as diabetes duration. Hence, we showed the importance of early and sustained good glycaemic control to prevent premature CVD in people with type 1 diabetes.

In Chapter V, we showed that higher levels of PA and less sedentary behaviour were associated with lower arterial stiffness, and VO_2max independently associated with cf-PWV in the multivariate model of Chapter IV. Further, PA was favourably associated with long-term glycaemic control, indices of body composition, and estimated insulin sensitivity. Engagement in regular PA and limiting sedentary behaviour could be important to impact metabolic and CV health in this population. We highlighted that future longitudinal studies should explore the interactions between and possible synergistic effects of PA, body composition and insulin sensitivity, and the possible effects of PA on hard CV endpoints by lowering arterial stiffness.

The third aim was to enhance our understanding of the glycaemic effects of postprandial exercise in adults with type 1 diabetes.

In Chapter VI, we provided the first comprehensive overview of the glycaemic effects of postprandial exercise in type 1 diabetes, including twenty studies investigating various exercise modalities. It was shown that blood glucose declines during all modalities (absolute declines between 2–5 mmol/L for CONT MOD, 3–8 mmol/L for CONT HIGH, 2–4 mmol/L for IHE and 2–3 mmol/L for HIIT, being clinically important decreases), seemingly regardless of the magnitude of pre-exercise bolus insulin reduction but depending on exercise duration and intensity. Therefore, we recommended that individuals with type 1 diabetes exercising postprandially should substantially reduce insulin bolus with the pre-exercise meal to avoid exercise-induced hypoglycaemia, with the magnitude of the reduction likewise depending on exercise duration and intensity. We also indicated that pre-exercise blood glucose and timing of exercise should be considered to avoid hyperglycaemia around exercise. Still, caution for late-onset post-exercise hypoglycaemia in the hours and overnight period is warranted.

In Chapter VII, we conducted a randomized controlled crossover trial (CARDIA 2.1-study) to further gain insight into the precise impact of postprandial exercise and thereby improve glycaemia management around activity (before, during and after). We investigated the effects of either rest, walking, continuous, or interval exercise performed one hour after the meal, on blood glucose as well as on CGM-parameters during and up to 24 hours after exercise. We found that all postprandial exercise modalities were effective, safe and feasible if necessary precautions are taken (i.e., prandial insulin reductions), as exercise lowered maximal postprandial glucose excursions and caused a consistent and clinically relevant blood glucose decline during exercise while there was no hypoglycaemia during or shortly after exercise. However, we highlighted two important issues we believe are remaining challenges with postprandial exercise: subacute post-exercise hyperglycaemia and nocturnal hypoglycaemia.

Altogether, the findings derived from this thesis are important steps to contribute to improving diabetes control and CV health, thereby ameliorating quality of life and reducing risk of premature CV morbidity and mortality in people with type 1 diabetes.