Trajectory of clinical parameters and incidence of cardiovascular disease in patients with type 2 diabetes mellitus

Abstract

The prevalence of Type 2 diabetes mellitus (T2DM) was increasing over decades. Cardiovascular disease (CVD) is the leading cause of mortality and affects approximately one-third of T2DM patients. CVD risk in T2DM patients is associated with clinical parameters including haemoglobin A1c (HbA1c), systolic blood pressure (SBP) and low-density lipoprotein cholesterol (LDL-C). The trajectories of these clinical parameters could be independent risk factors of CVD in T2DM patients in addition to single baseline measurements. Therefore, this thesis aimed to investigate the association between the trajectory of clinical parameters and the concurrent CVD incidence in T2DM patients, which had been poorly understood. This was a retrospective cohort study conducted in Hong Kong. This study used the data from the Clinical Management System of the Hospital Authority (HA). Adults with T2DM who were managed in the HA were included. A trend analysis investigated the age-standardised trends of clinical parameters, T2DM complications and mortality of a fixed cohort. In each of the six subgroups defined by baseline value of clinical parameter (low/medium/high) and age (young/old), the association between the trajectory of each clinical parameter and the concurrent CVD incidence was investigated through clustering patients into latent classes jointly by the longitudinal trajectory and survival patterns using the Joint Latent Class Mixed Model, adjusting for baseline value of clinical parameters. The mean predicted trajectory of HbA1c, SBP and LDL-C and the estimated cumulative incidence of CVD were explored. Multinomial Logistic Regression was used to investigate the association between baseline characteristics and the membership of latent classes. A total of 259,211 T2DM patients were included in this study. The trend analysis observed decreasing LDL-C and annual incidences of most T2DM complications, including CVD, but increasing annual incidences of renal complications and mortality. Younger patients had increasing incidences for all T2DM complications and mortality. Most patients were found to have stable trajectories of HbA1c and SBP and decreasing trajectories of LDL-C. Patients with fluctuating HbA1c trajectories and those with fluctuating or increasing SBP trajectories had higher CVD incidence compared with those with stable trajectories. In patients with higher baseline LDL-C, fluctuating LDL-C and decreasing LDL-C in a later stage were associated with higher CVD incidence compared with decreasing LDL-C in an early stage. Rapid decreases from high baseline values of HbA1c, SBP and LDL-C were accompanied with higher CVD incidence. The association between baseline characteristics and trajectory patterns varied between younger and older subgroups. The trajectory of clinical parameters is associated with CVD incidence in T2DM patients. Fluctuations in HbA1c, SBP and LDL-C need more attention in T2DM management. Timely lipid-lowering treatment is important in patients with high LDL-C to reduce CVD risk. The management of T2DM patients should be individualized by younger and older patients to address the potential pathology or lifestyle heterogeneities. The poorer trends in annual incidence of complications and fluctuations in clinical parameters in younger T2DM patients underscore the need of closer monitoring and tighter control of risk factors. Patients with intensive treatment may need more attention in case of possible adverse effect.