In this prospective cohort of young adults followed up over 25 years, we demonstrated greater IGE during young adulthood was associated with unfavorable impairment of multiple target organs, including subclinical cardiac structural and functional impairment, subclinical atherosclerosis and albuminuria independent of fasting glucose at Y25. The long-term IGE was associated with prevalence of TOD and the higher IGE tended to involve more target organs, which remained consistent even among those free from diabetes at Y25.
Previous researches have established that diabetes is responsible for multiple target organs injury . Diabetic cardiomyopathy is characterized by myocardial hypertrophy and myocardial remodeling, manifested as diastolic dysfunction in the earlier and systolic dysfunction during disease progression . Besides, individuals with diabetes are also at increased risk of atherosclerosis and diabetic nephropathy in later life [3, 20]. Overall, these prior studies suggested that diabetes is a strong predictor of adverse clinical events in later life. However, previous studies always investigated the detrimental impacts of hyperglycemia based on measurement of blood glucose at a single time point and whether lifespan exposure to hyperglycemia derived subclinical adverse effects earlier is still unknown. To the best of our knowledge, our study was the first to comprehensively focus on the chronic impacts of long-term intensity of glycemic exposure during young adulthood on multiple target organs in midlife, accounting for the magnitude and duration of glycemic exposure simultaneously. Our findings extended the prior studies by showing that higher IGE over 2nd, 3rd and 4th decade of life was associated with earlier target organs dysfunction at 5th decade. Besides, our results also suggested that the numbers of target organs involvement depended on the cumulative effects of IGE, even in a non-diabetic subset of participants who were younger with less clinical comorbidities.
The underlying pathophysiologic pathways that links higher long-term IGE with target organs impairment may share common pathogenic mechanisms. Study have shown that hyperglycemia induces and accelerates the deposition and accumulation of glycosylation products within myocardium and arterial wall, leading to cardiac and vascular injury [19, 21, 22]. Besides, endothelial dysfunction is also a main pathogenic factor contributing to diabetic vascular impairment, e.g., initiating the pathogenesis of atherosclerosis . Furthermore, hyperglycemia and insulin resistance also promote myocardial collagen deposition and myocyte hypertrophy, causing myocardial remodeling . Some other mechanisms may also involve in the pathogenesis, including oxidative stress, altered substrate metabolism, mitochondrial dysfunction, inflammation activation and so on [19, 24, 25].
For cardiac functional measurements, it has been demonstrated that global longitudinal strain is a sensitive marker of systolic dysfunction than ejection fraction and provide additional prognostic value on adverse outcomes [12, 26]. Meanwhile, previous studies indicated that E/e’ ratio was more predictive for primary cardiac events than E/A ratio . In our study, we noted associations between long-term IGE with worse global longitudinal strain but not ejection fraction. We also found that IGE over young adulthood was associated with worse E/e’ ratio rather than E/A. Therefore, GLS and E/e’ ratio seemed to be powerful in showing the adverse effects of IGE that classical echocardiographic indicators failed to identify, which may be explained by a young age of CARDIA population with lower prevalence of clinical organ dysfunction at advanced stage. Similar, IGE was also associated with CAC and albuminuria, the well-known and powerful precursors of adverse outcomes [5, 28]. In fact, identification of the predisposing factors for subclinical organ dysfunction provide the insight towards long-term risk of organ dysfunction. From a disease-prevention perspective, given the independent prognostic significance of TOD, it would make sense to establish screening strategies on IGE and implement intensive blood glucose management during young adulthood, avoiding irreversible organs damage [4–6].
An unexpected finding was that long-term IGE showed subclinical adverse effects on target organs in both diabetes and non-diabetes subgroups, which highlights the need to investigate the cumulative effects of glycemic exposure in participants below the diagnostic criteria of diabetes. Indeed, even in examination at Y25, only a minority of participants (13.0%) were diagnosed as diabetes and the FG level in non-diabetes population was on average below the current threshold of prediabetes (92.8 ± 9.1 mg/dl, Supplemental table 2), suggesting exposure to subclinical hyperglycemia over young adulthood could precipitate organ dysfunction. Prospective study may be warrant to determine the optimal cut-off points for evaluation of the subclinical impacts of cumulative glycemic exposure. On the other hand, given the subclinical hyperglycemia is a modifiable factor, the optimal glycemic control target remains controversial despite several randomized controlled trials (RCTs) published [29–32]. UKPDS trial demonstrated a great microvascular benefit from intensive glucose control therapy, which was confirmed in subsequent ADVANCE and ACCORD trial. However, whether intensive glucose control therapy reduce risk of macrovascular events is still disputed. Indeed, the participants in these RCTs were characterized by old age, long duration of diabetes and poor glycemic control. Irreversible subclinical organ impairment in these participants may weaken the effects of intensive glucose control therapy and lead to unclear macrovascular benefit. Thus, further research is need to clarify whether intensive glucose control in young patients with subclinical hyperglycemia can prevent or delay the early organ dysfunction, ultimately improving prognosis. Nevertheless, the accompanying risk of adverse reaction, especially hypoglycemic episodes, should be carefully assessed.
Our study results are strengthened by a population-based cohort with large sample size; regular screening fasting glucose using standardized protocols over 25 years, facilitating assessment of cumulative intensity of glycemic exposure; a cohort of young adults with less comorbidities, almost non-diabetic population, was optimal for exploring the subclinical effects of IGE on target organs and providing evidences on prevention of diabetes-induced end-organ damage. However, several limitations in our study needs to be considered. First, cardiac function and CAC were measured in detail until examination in Y25, hindering further assessment of the long-term changes of target organs function from young adulthood to midlife. Hence, we can only assume that the measurements in Y25 could reflect the declines of organs function. Second, only approximately a half of participants in CARDIA performed echocardiographic measurements and CAC assessment in Y25, thus our analysis may be subjected to selection bias. To address this issue, we further compared the baseline characteristics between the analyzed sample with the excluded sample (Supplemental table 1). The participants who were excluded in our study were more frequently male, black, smoker, have lower educational attainment, higher FG and SBP levels. In fact, the participants included in our analysis may have a lower risk of target organs impairment than those excluded, which likely bias towards the null and may underestimate the deleterious effects of IGE. Third, owing to the calculation method of IGE, each individual had unequal number of measurements of FG and those who were less measured may have inaccurate evaluation of glycemic exposure. Indeed, more than 80% of individuals completed all 6 measurements of FG and we observed consistent associations when additionally adjusted for number of measurements (Supplemental table 3). Fourth, given the nature of observational study, our findings may be susceptible to reverse causation and residual confounding. Last but not least, CARDIA is a biracial cohort including white and black individuals so that our findings required external validation in other ethnic population, e.g., Asian.