Insulin resistance predicts progression of de novo atherosclerotic plaques in patients with coronary heart disease: a one-year follow-up study

Although drug-eluting stents prevail in treating coronary heart disease (CHD), several concerns have already begun to rise publicly, including those regarding medical complications social burdens [1]. More cost-effective preventive treatments are called for; therefore, various studies have been conducted to identify the risk factors for CHD, which are classified into classical versus novel categories [2]. The former consists of aging, female sex, genetic factors, obesity, smoking, dyslipidemia, hypertension and diabetes mellitus while the latter includes several inflammatory markers such as hsCRP, IL-6, TNF-α, PAI-1, MCP-1 and adhesion factors such as ICAM-1 and VCAM-1 [2]. Some studies also divert their targets on the specific group of patients such as patients with hypertension, diabetes, metabolic syndrome or even SLE to personalize the existing grading systems of risk factors including the classic Framingham, Reynolds, SCORE and ASSIGN [2]. Insulin resistance, frequently appears in various clinical settings such as hypertension, diabetes and metabolic syndrome and is believed to be responsible for connecting endocrinological disorders with their potential adverse cardiovascular complications based on its reciprocal relationship with endothelial dysfunction indicated by cellular, physiological, clinical, and epidemiological studies [3–5]. However, few studies have actually examined its relationship with the progression of coronary atherosclerotic plaques. The aim of our study was to explore the effect of insulin resistance on the progression atherosclerotic plaques in patients with CHD.

Since June 7^th, 2007 to Dec 30^th, 2011, subjects were identified and screened from patients seeking a planned percutaneous intervention for suspected new onset of CHD based on the clinical presentation, which was categorized as ECG abnormality (only positive stress ECG), stable angina (stable exertional symptoms only), unstable angina (progressive symptoms or symptoms at rest), or MI (creatinine kinase [CK] level >636 IU/L and creatinine kinase-myocardial band isoenzyme [CK-MB] index >6%) in the catheter Lab at the Shanghai Ruijin Hospital which is affiliated with the Shanghai Jiaotong University School of Medicine . Written consent was obtained from all individuals before they were enrolled in the study. Their medical histories were obtained and recorded, and pertinent baseline laboratory studies were drawn before cardiac catheterization while BMI and GFR were calculated. All patients were treated with IV heparin and combined antiplatelet therapy, while the use of IIb/IIIa inhibitors was at the operator’s discretion. Telephone interviews were performed at 6 months and participants were readmitted for routine angiographic follow-up 1 year after the procedure, including those with normal angiography at their index visits. Necessary cardiac catheterization was performed for recurrent symptoms or objective evidence of ischemia. Individuals were excluded from the study if they fell into one of the following criteria: Patients with histories of CHD or prior coronary revascularization; patients undergoing CABG instead of stenting; patients with heart failure or any kind of cardiomyopathy; patients with familial hypercholesterolemia; patients with severely impaired liver or renal function; patients with a terminal illness with a life expectancy of less than 1 year.

Coronary angiography and Gensini score

Standard method was adopted and Judkins method was deployed. All patients were admitted to the hospital the day before CAG. The purpose of CAG was to diagnose any ischemic conditions such as angina. After administration of isosorbide dinitrate (2.5–5 mg bolus dose), the coronary arteries were evaluated by 2 experienced cardiologists. Narrowing of ≥50% in one of the three major coronary arteries or their major branches was judged significant while narrowing of <5% was considered insignificant. We defined both residual narrowing of less than 20% and third class TIMI blood flow as successful outcomes of the procedure. The decision to deploy a balloon or use a specific drug-eluting stent was made by the cardiologists themselves. All angiograms were evaluated in similar angiographic angles, and the percentage of stenosis was determined with the use of calipers comparing the stenotic segment with the proximal, angiographically normal segment. Clodigrel, 75 mg per day, or ticlopidine, 250 mg twice per day, was administered for at least 12 months after the procedure and all patients received 100 mg aspirin per day. The same guidelines were implemented at 1-year follow-up for each subject. We also adopted the modified Gensini scoring system to evaluate both the baseline and the follow-up angiograms [10]. To summarize this scoring system, five points were given for left main lesion; 2.5 points for the proximal left anterior descending artery (LAD) or left circumflex (LCX); 1.5 points for mid-segment LAD; one point for the distal segment of LAD, first diagonal branch, LCX obtuse marginal branch or right coronary artery, and 0.5 points for the second diagonal branch or LCX posterolateral branch (Figure 1). To define the atherosclerotic plaque progression, we divided the following patients into the progression group: (1) patients who progressed from normal or insignificant angiogram to CAD; (2) Patients whose documented baseline lesions exacerbated in the same vessel; (3) Patients who progressed from their baseline CAD to new lesion(s) in a different vessel in one-year period. In addition, we classified those who had normal angiograms or similar CAD lesions at the first and following visits into the non-progression group [11].

Our study explored the effect of insulin resistance on the progression of atherosclerotic plaques over a 1-year follow-up interval. The difference value of Gensini score between the higher IR group and the lower IR group was statistically significant and the binomial multivariate regression models contended that insulin resistance was an independent predictor of atherosclerotic progression in patients with coronary heart disease irrespective of the individual’s diabetes status. Insulin resistance generally occurs in cluster with other risk factors including hyperglycemia, dyslipidemia and inflammatory conditions. IR itself is a phenomenon that decreases the effects produced by the normal activity of insulin [13]. In addition, one of its key biochemical defects lies in the specific impairment of PI3K-dependent signaling pathways while other pathways including MAPK were intact [14], which results in the overproduction of ET-1 over eNOS, leading to endothelium dysfunction and enhanced levels of VCAM-1 and MCP-1, eventually contributing to the formation of atherosclerotic plaques [15]. Early studies have already supported the idea that IR is an important predictor of CHD [16]. Tetsuya conducted a prospective study and found that IR is associated with coronary lipid-rich plaques in patients with abnormal glucose regulation [17]. Reports have also indicated that IL-6, a potential cardiometabolic biomarker expressing in human atherosclerotic lesions, has a close relationship with HOMA-IR [18]. Uli C argues that low adiponectin levels are frequently associated with insulin resistance, may have a closer link with coronary plaque vulnerability, which plays an important role in the pathogenesis of ACS [19]. Having enrolled 543 patients with diabetes and CHD and using the intravascular ultrasound (IVUS) technique to assess the progression of coronary atherosclerosis directly, the recent PERISCOPE study concluded that pioglitazone, a drug targeting IR, could significantly slow the progression of atherosclerosis, which is also supportive of our study [20].

Having confirmed the predictive value of IR on plaques progression, our study also looks into the issue of in-stent restenosis, an important complication of coronary arterial intervention nowadays [21]. Having observed similar prevalence of in-stent restenosis between the higher and lower IR groups, we set up the binary logistic regression model to evaluate the risk factors for in-stent restenosis by inputting the same variables as above. The results showed that unlike prevalence of HTN (OR = 0.241, p = 0.019) and age (OR = 1.084, p = 0.008), insulin resistance had no statistically significant effect on the development of in-stent restenosis (OR = 2.064, p = 0.218, table not shown). It has been well studied that the formation of in-stent restenosis depends on the proliferation of in-stent neointimal cells because of complex inflammatory responses in the short time interval, which is quite different from the process of plaque formation [21, 22]. This may serve to explain the results of our study. However, conflicting evidence does exist. A study in 2005 by Kazuaki stated that IR predicted in-stent restenosis rather than de-novo stenosis, which was mediated by atherosclerosis plaque progression [7]. Following the difference can be utilized to explain the differences in results between their study and ours according to the following facts: 1. the aim of the study by Kazuaki was to evaluate the IR’s effect on restenosis after stenting, which resulted in the 110 recruited individuals were not just confined to subjects with CHD. Instead, the sample included patients who had extensive coronary artery dissection after PTCA, complete vessel closure, residual stenosis of 25% or more of the vessel diameter. 2. The follow-up interval in Kazuaki’s study, which played an important role in the development of in-stent restenosis was 4-months. In our case, the interval was 1 year. 3. Instead of setting up a cutting value for HOMA-IR to determine the state of insulin resistance as we did, they set the HOMA-IR as a continuous variable. 4. The variables they used in the multivariable logistic regression model only included HbA1C, HOMA-IR and LDL-C, which in our case are risk factors such as age, sex, BMI, prevalence of hypertension and diabetes, HOMA-IR, HbA1c, hsCRP, LDL-C and urine MA/Cr. Based on these differences in the fundamental aspects of the studies, it seems inappropriate to compare their results with ours.

Our study also revealed that HbA1c also stands for an independent predictor of coronary plaque progression in unselected individuals. As a reliable marker of recent 2–3 months serum glucose level, HbA1c has already been used for diagnosing diabetes mellitus [23, 24]. In a recent meta-analysis covering 20 studies involving 13, 224 individuals suggested that HbA1c level was an independent risk factor for mortality in CHD patients without diabetes [25]. Hiroyasu also stated that HbA1c was an independent predictor of major adverse cardiac events after the implantation of DES [26]. Several other studies also support the role of HbA1c in future CVD event and mortalities [27–29].

Insulin resistance is an independent predictor for atherosclerosis plaque progression in patients with coronary heart disease in both the diabetic and non-diabetic population.