We first conducted a case–control study to evaluate the effect of metabolic factors on DHF in Chinese high-risk CAD patients. CAD, HT and DM were more prevalent in patients with MetS. Most of the demographic factors, biochemical characteristics and echocardiographic measurements significantly differed among the three MetS groups. Doppler echocardiography has become a well accepted, reliable, noninvasive tool to measure the LV diastolic function. In the present study, Doppler echocardiography was used to measure LV diastolic function in order to diagnose DHF.
The main finding of the present study was that MetS strongly and independently predicted DHF in high-risk CAD patients. The prevalence of DHF increased with the severity of MetS. HT, insulin resistance and obesity have been associated with LV diastolic dysfunction or DHF in different populations [11, 12]. In addition, MetS has been independently correlated with DHF in different subgroups such diabetic, non-diabetic or hypertensive patients [13–15]. In the present study, the association between MetS and DHF was observed in both the univariate and multivariate models after adjustment for potential confounders in high-risk CAD patients. Specifically, we found a good association between the MSSs and DHF. To our knowledge, this is the first study to have reported such an association in a population of high-risk CAD patients. In the multivariate analysis, MetS was independently associated with DHF, even after adjustment for potential confounders such as parameters of renal function, LVMI and CAD. This finding is of special importance if the direct relationship between MetS and DHF is considered. The clustering of cardiovascular risk factors in MetS indicates that the multiple complex metabolic reactions involved in glycotoxicity, lipotoxicity, altered insulin signaling, increased cytokine activity and interstitial deposition of triacylglycerol may directly or indirectly impact myocardial function [4, 16–20]. Additionally, these metabolic risk factors lead to reduced energy availability, and have an additive, adverse effect on endothelial function .
In the present study, the AUC was calculated to show that the MSS strongly predicts DHF. In patients with MSSs of up to 4, the prevalence of DHF was nearly 90%. This finding indicates that the severity of MetS is linked to the progression of DHF. This is one of interesting findings of the present study, which not only supports further studies on the mechanism of DHF but also provides evidence for clinicians to predict DHF in hospitalized patients. However, in the present study, we scored the MetS severity by simply using the number of MetS criteria. We did not consider the weights of the MetS components. For instance, the BP component of MetS makes a greater contribution to DHF. A large-scale, case–control study or cohort study with a better method of scoring MetS severity will be conducted to develop a highly sensitive and specific model that uses MetS information to predict DHF. Such a model would facilitate the prevention and treatment of DHF in clinical practice.
Another interesting finding of the present study was that BP and TG were the only MetS components that contributed to DHF. This finding is inconsistent with those of some earlier studies, which had revealed that BMI, SBP, DBP and lipid profiles were significantly associated with diastolic parameters and the structure and functions of the LV [4, 16–18, 20]. In the present study, BMI, FPG and HDL were not significantly associated with DHF. This difference is partly because the contributions of individual MetS components could not be detected in the present study, which had a moderate sample size. Another possible cause is that the present study population differed from those in previous studies; we performed association analysis for MetS and DHF in high-risk CAD patients with HT, DM, CAD and hyperlipidemia, which were potential confounders of DHF. In addition, MetS components as continuous variables may not reflect the true values in patients undergoing medical therapy. For example, the FPG values measured in the present study were less than the true values in DM patients using hypoglycemic drugs. However, we focused on metabolic factors associated with DHF in a specific subgroup, high-risk CAD patients. Our findings will provide evidence for clinicians to better understand and treat patients in this specific subgroup. Nevertheless, further studies should explore the effects of MetS on DHF in an exclusive subgroup such as patients with CAD or hyperlipidemia. DBP as a continuous variable was found to be associated with DHF in both the univariate and multivariate models. The results were confirmed in the MLR analysis with BP as a binary variable. DBP has been found to be an important predictor of LV diastolic dysfunction or DHF [20, 22]. DBP can directly influence diastolic function and remodel the LV structure, leading to DHF . In the present study, TG as a binary variable was associated with DHF in both the univariate and multivariate analyses. Other studies have reported similar results [11, 24]. No consistent results have been found in MLR analyses with TG as a continuous variable. This is partly because high-risk CAD patients were regularly treated with anti-lipids drugs such as statins to prevent events of cardiovascular disease; this may have influenced the true value of TG, making it difficult to determine the effect of TG on DHF. The exact mechanism underlying the association between TG and DHF has not been fully elucidated. In the present study, we did not determine the mechanism via which TG modifies metabolic factors and induces DHF.
Several limitations of the study deserve comment. First, the design of the present study was hospital-based, which is susceptible to selection bias. Second, the sample size was moderate, limiting its ability to detect significant results. Third, the multiple regression models indicated only a moderate influence of MetS on DHF. Other environmental and genetic factors may contribute to the unexplained variation in DHF prevalence. Fourth, the association between insulin resistance and DHF was not analyzed in the present study. This is because data on fasting blood insulin levels were missing. Furthermore, most of participants were enrolled with the first diagnosis of diastolic heart failure or not. So we did not collect information of the history of diastolic heart failure. Finally, it is important to mention that our study was performed on Chinese individuals, and our findings may not be relevant to people of other ethnicities.