Glycemic Control, Cardiac Autoimmunity, and Long-Term Risk of Cardiovascular Disease in Type 1 Diabetes Mellitus

Abstract

Background: Poor glycemic control is associated with increased risk of cardiovascular disease (CVD) in type 1 diabetes mellitus (T1DM); however, little is known about mechanisms specific to T1DM. In T1DM, myocardial injury can induce persistent cardiac autoimmunity. Chronic hyperglycemia causes myocardial injury, raising the possibility that hyperglycemia-induced cardiac autoimmunity could contribute to long-term CVD complications in T1DM.

Methods: We measured the prevalence and profiles of cardiac autoantibodies (AAbs) in longitudinal samples from the DCCT (Diabetes Control and Complications Trial) in participants with mean hemoglobin A_1c (HbA_1c) ≥9.0% (n=83) and ≤7.0% (n=83) during DCCT. We assessed subsequent coronary artery calcification (measured once during years 7-9 in the post-DCCT EDIC [Epidemiology of Diabetes Interventions and Complications] observational study), high-sensitivity C-reactive protein (measured during EDIC years 4-6), and CVD events (defined as nonfatal myocardial infarction, stroke, death resulting from CVD, heart failure, or coronary artery bypass graft) over a 26-year median follow-up. Cardiac AAbs were also measured in matched patients with type 2 diabetes mellitus with HbA_1c ≥9.0% (n=70) and ≤7.0% (n=140) and, as a control for cardiac autoimmunity, patients with Chagas cardiomyopathy (n=51).

Results: Apart from HbA_1c levels, the DCCT groups shared similar CVD risk factors at the beginning and end of DCCT. The DCCT HbA_1c ≥9.0% group showed markedly higher cardiac AAb levels than the HbA_1c ≤7.0% group during DCCT, with a progressive increase and decrease in AAb levels over time in the 2 groups, respectively ( P<0.001). In the HbA_1c ≥9.0% group, 46%, 22%, and 11% tested positive for ≥1, ≥2, and ≥3 different cardiac AAb types, respectively, similar to patients with Chagas cardiomyopathy, compared with 2%, 1%, and 0% in the HbA_1c ≤7.0% group. Glycemic control was not associated with AAb prevalence in type 2 diabetes mellitus. Positivity for ≥2 AAbs during DCCT was associated with increased risk of CVD events (4 of 6; hazard ratio, 16.1; 95% CI, 3.0-88.2) and, in multivariable analyses, with detectable coronary artery calcification (13 of 31; odds ratio, 60.1; 95% CI, 8.4-410.0). Patients with ≥2 AAbs subsequently also showed elevated high-sensitivity C-reactive protein levels (6.0 mg/L versus 1.4 mg/L in patients with ≤1 AAbs; P=0.003).

Conclusions: Poor glycemic control is associated with cardiac autoimmunity in T1DM. Furthermore, cardiac AAb positivity is associated with an increased risk of CVD decades later, suggesting a role for autoimmune mechanisms in the development of CVD in T1DM, possibly through inflammatory pathways.

Keywords: autoantibodies; biomarkers; cardiovascular diseases; diabetes mellitus; hyperglycemia.

Figure 1.. Levels of cardiac autoantibodies in subjects from the different groups.

AAb, autoantibody; A1c, glycosylated hemoglobin level; dotted lines, 99^th percentile cutoffs for AAb positivity. AAb levels for the T1D patients are the mean AAb index during DCCT. S1-MYH6, S1 fragment of cardiac α-myosin heavy chain (MYH6); S2-MYH6, S2 fragment of MYH6; cTnI, cardiac troponin I; FL-MYH6, full-length-MYH6; and FL-MYH7, full-length β myosin heavy chain. Controls, healthy control subjects. Inset: Subjects who subsequently had CVD events during EDIC. Number below “Masked_ID” is the DCCT/EDIC masked identifier (ID) number; CVD, cardiovascular disease; MI, nonfatal myocardial infarction; CABG, coronary artery bypass graft; NA, not applicable.

Figure 2.. Prevalence of subjects having ≥1, ≥2, ≥3, and ≥4 different types of cardiac autoantibodies (AAb) in the different study groups.

The groups include: a) type 1 diabetes DCCT participants with time-updated mean HbA1c levels of ≤7.0% or ≥9.0% during DCCT, b) patients with chronic Chagas cardiomyopathy (Chagas cardiomyopathy); c) type 2 diabetes patients with HbA1c ≤7.0% and HbA1c ≥9.0%; and d) healthy control subjects (‘Controls’). Number of AAb includes positivity to S1-MYH6, S2-MYH6, cTnI, FL-MYH6 and/or FL-MYH7.

Figure 3.. Relationship between autoantibody level (AAb), HbA1c group, year of DCCT treatment, and time-updated mean HbA1c levels (mean HbA1c levels).

Panel A. Interaction between the mean AAb level to S1-MYH6 (top panel), S2-MYH6 (middle panel) or FL-MYH6 (lower panel), HbA1c group (≥9.0% or ≤7.0%), and year of DCCT intervention treatment. All relationships were significant (P<0.001). Because only 2 subjects had samples available from DCCT Year 0 (i.e., DCCT baseline), Year 0 is not included. The numbers of samples analyzed in each HbA1c group per DCCT year are as indicated. Insets show the patients who developed CVD events (as shown in Figure 1). Panel B. Time-updated mean HbA1c levels in each of the HbA1c groups per year of DCCT treatment. Cutoffs for AAb positivity are: S1-MYH6, 0.330; S2-MYH6, 0.763; and FL-MYH6, 0.701.

Figure 4.. Relationship between the HbA1c groups (left panel) or number of cardiac autoantibodies (right panel) during DCCT and hsCRP levels during EDIC years 4–6.

hsCRP, high-sensitivity C-reactive protein. Gray bars indicate the medians in the different groups and dotted lines indicate the upper reference limit for hsCRP (3.0 mg/L). P values refer to the comparison of log-transformed values.

Figure 5.. Proposed scheme of how chronic hyperglycemia is associated with cardiac autoimmunity and increased risk of CVD in patients with type 1 diabetes.

A) Absent thymic expression of the α-isoform of myosin heavy chain (α-myosin, encoded by MYH6) is associated with high frequencies of CD4⁺ T cells specific to α-myosin in the peripheral blood of individuals in the general population; B) In both type 1 and type 2 diabetes (T2D) patients, chronic hyperglycemia causes subclinical myocardial injury, leading to leakage and exposure of heart muscle proteins, including α-myosin, to the immune system; C) In T1D patients with poor glycemic control, the dysregulated adaptive immune system overreacts to myocardial injury, leading to the expansion of proinflammatory CD4⁺ T cells specific to α-myosin and the development of AAb to MYH6 and other cardiac antigens. This proinflammatory state is associated with elevated levels of the inflammatory marker, high-sensitivity C-reactive protein (hsCRP), and increased risk for accelerated atherosclerosis and CVD events.