Observational studies in a diabetic population: The impact of glycemic control on the risk of fractures and venous thromboembolism

Diabetes Mellitus (DM) is an increasingly prevalent chronic condition affecting around 450 million patients worldwide in 2017. It is estimated to cause around 5 million deaths per year. The main types of diabetes are type 1 (T1DM) and type 2 diabetes mellitus (T2DM). Both types of DM are associated with an increased risk for several diseases and health problems, including fractures and venous thromboembolisms (VTEs). While some diseases and health problems have been directly and causally linked to diabetes, the association between DM and other diseases is unclear. This is, amongst others, the case for the association between DM and fractures, as wells as between DM and VTE. Since DM is a disease characterized by hyperglycemia over prolonged periods of time, the level of glycemic control is a key aspect of DM treatment. With its proper control, the onset of several complications can be delayed or prevented. This raises the question of whether good glycemic control could also impact the association between DM and diseases with an unclear direct causal link to DM, such as fractures and VTEs.
Thus, this thesis aimed to contribute to the general understanding of how the level of glycemic control in patients with T1DM and T2DM may affect DM complications, more specifically the risk for fractures and VTE. To shed light onto this matter, we conducted three exemplary studies aiming at answering three different research questions. The studies were performed in specifically defined study populations with data taken from the Clinical Practice Research Datalink (CPRD) GOLD.
In study I, we comprehensively assessed the association between the level of glycemic control and the risk of low-trauma fractures in patients with T1DM and T2DM. Patients with newly diagnosed T1DM and T2DM between 1995 to 2015 were included in this 1:4 matched case-control study and we used conditional logistic regression to perform the analyses. The study included over 3’200 patients with T1DM and over 44’000 patients with T2DM and allowed us to have a detailed insight into the characteristics of the DM population in the UK. We found the median duration between the onset of T1DM or T2DM and the low-trauma fracture event to be 4.5 years, meaning that they were identical for both types of DM. The T1DM population included 46% female patients, the T2DM population, however, included over 71% female patients. While the risk of fracture was increased in patient with T1DM with mean hemoglobin A1c (HbA1c) level of > 8.0% (aOR 1.39, 95% CI 1,06-1.83) when compared to those with T1DM and mean HbA1c levels ≤7.0%, we saw no such effect in patients with T2DM. In other words, the level of glycemic control had no impact on the risk of fracture in the T2DM population. Accordingly, we found that the effect of glycemic control on the risk of low-trauma fracture differs between patients with T1DM and T2DM. This shows that more studies assessing the impact of glycemic control in patients T1DM, but especially in those with T2DM, are necessary to explore this difference in detail.
Building on our previous results, in study II we deepened the understanding of the association between the level of glycemic control, the use of antidiabetic medication, and the risk of low-trauma fractures in patients with T2DM. We included patients with newly diagnosed T2DM from 1995 to 2017 into our 1:4 matched case-control study. Our exposures of interest were glycemic control (measured through HbA1c levels) and the antidiabetic medication schemes. We identified almost 9000 cases and were able to assess the association between glycemic control and the risk of low-trauma fracture in patients receiving the following medication schemes: No drug treatment (only behavioral and dietary recommendations), metformin monotherapy (initial drug treatment), metformin plus either DPP4-inhibitors, glitazones, or sulfonylureas (first intensification of drug treatment), and metformin plus 2 drugs out of DPP4-inhibitors, glitazones, or sulfonylureas as well as any treatment regimen containing insulin (second intensification treatment). Like in our previous study, most patients included in the study were female (70%). We found that patients with current use of metformin and HbA1c levels <7.0% and between 7.0-8.0% had a reduced risk of fractures (aOR 0.89, 95% CI 0.83-0.96 and 0.81, 95% CI 0.73-0.90, respectively) compared to patients receiving no drug treatment. However, we did not find an association between good glycemic control and the risk of low-trauma fractures in patients receiving other antidiabetic treatment schemes when compared to medically untreated patients. This study highlighted the need for a better understanding of the interaction between the different involved characteristics of DM and its treatment and their association to the risk of low-trauma fractures.
In study III we directed our focus toward the association between glycemic control (measured as HbA1c levels) and the risk of unprovoked VTE in patients with T2DM. We assessed this in a 1:4 matched, nested case-control study within a cohort of patients with newly diagnosed T2DM between 1995 and 2019. We identified 2’653 VTE cases, out of which 53.1% were female. To ensure our cases had an unprovoked VTE, we excluded patients with a history of VTE (at any time prior to the diagnosis of T2DM), a code for surgery, immobilization, trauma, paralysis and paresis, or a code for the use of hormone replacement therapy or the contraceptive pill within 3 months prior to the VTE. We also excluded patients with a code for pregnancy of puerperium within 12 months prior to the VTE. We assessed the level of glycemic control using 7 categories: ≤6.5%, >6.5-7.0% (reference group), >7.0-7.5%, >7.5-8.0%, >8.0-9.0%, >9.0%, and no HbA1c measurement. We found no association between the last HbA1c measurement and the risk of VTE in patients with T2DM. However, in an analysis taking only the HbA1c value within 90 days prior to the VTE into account, women with HbA1c levels >7.0% had a 36-55% increased relative risk of VTE when compared to women with HbA1c levels >6.5-7-0%. With this result, our study raised the suspicion that female T2DM with HbA1c levels >7.0% may have a slightly higher risk for unprovoked VTE when compared to women with HbA1c levels >6.5-7.0%. We observed no similar effect of glycemic control on the risk of VTE in men.
In summary, the studies presented in this thesis contribute to the understanding of the impact of glycemic control on two important health conditions that are more common in the diabetic population than in those without diabetes: low-trauma fractures and VTE. While this thesis provided answers to research questions related to the association between glycemic control and the risk of fractures and VTE, the need for further research in DM and its complication is obvious, especially given that the number of patients with DM worldwide is steadily growing.