Asymptomatic coronary artery disease in a Norwegian cohort with type 2 diabetes: a prospective angiographic study with intravascular ultrasound evaluation

This is the primary IVUS report of the Asker and Baerum Cardiovascular Diabetes (ABCD) study, and a detailed description of this study has been reported previously [13]. In brief, the ABCD study was a prospective, open, randomized, controlled study of 2 years of intensive versus standard care in 120 patients with T2D conducted at the Asker and Baerum Hospital, Gjettum, Norway. Patients were included during the period January 2002 to February 2004. Inclusion criteria were T2D, age 18–75 years, and the presence of one or more additional CV risk factor (defined as hypertension (treated or ambulatory systolic/diastolic blood pressure (BP) ≥ 140/90 mmHg), past or prior smoking, premature CAD in first degree family member (male < 55 years, female < 65 years), microalbuminuria or dyslipidemia (treated or total cholesterol ≥ 5.0 mmol/L, high-density lipoprotein (HDL)-cholesterol (< 1.0 mmol/L in men, or < 1.1 mmol/L in women or triglycerides ≥ 2.0 mmol/L).

Patients were randomized to 2 years of intensive, hospital-based, structured multi-intervention (n = 60) or standard care (n = 60). Structured intensive multi-intervention comprised 6 months of lifestyle intervention (i.e. advice on diet, exercise and smoking cessation and reimbursement of cost associated with training) followed by targeted, pharmacological therapy to reach pre-specified treatment goals (HbA1c ≤ 48 mmol/mol (6.5%), total/LDL cholesterol < 5.0/3.0 mmol/L, systolic/diastolic blood pressure (BP) < 130/80 mmHg). The participants were reviewed every 3 months over a period of 2 years by a diabetologist in the hospital’s out-patient clinic.

The standard care group remained under the care of their general practitioners who were encouraged to continue with treatment according to current (2002) national and American Diabetes Association [14] guidelines (HbA1c < 7%, LDL-cholesterol < 2.6 mmol/L, systolic/diastolic BP < 130/80) and follow-up recommended at 3-monthly intervals.

All participants underwent a comprehensive diagnostic work-up at baseline, including invasive coronary angiography regardless of symptoms or results of the non-invasive cardiac tests. Patients could refrain from invasive testing and still participate in the study, and 91 of the 120 enrolled consented to invasive coronary angiography. At 2 years follow-up, the diagnostic work-up at baseline was repeated, excluding invasive coronary angiography, and all participants were transferred back to the care they had prior to study entry. It was prespecified that the participants should enter a long-term follow-up, and approximately 7 years after inclusion, an additional diagnostic work-up was conducted, including coronary angiography supplemented with IVUS. Of the 120 patients included in the ABCD study, 85 participated in the long-term follow-up [15], and 56 (46.7%) of these patients had coronary angiography performed both at baseline and 7 years. These 56 patients (30 and 26 patients in the multi-intervention and control group, respectively) constitute the population of this sub-study. All 56 patients were free from cardiopulmonary symptoms at baseline.

The reference material consisted of IVUS of donor hearts from 147 non-T2D donors that a priori were free from symptomatic CAD, performed 7–11 weeks post transplantation.

Coronary angiography was performed with the percutaneous radial or femoral approach using 6F diagnostic catheters (Cordis Corporation, Miami, Fla., USA) and the water-soluble, non-ionic, dimeric contrast medium iodixanol (Visipaque 320 mg/mL; G.E.Healthcare, Oslo, Norway). Coronary artery angiogram data was evaluated by experienced local staff blinded to treatment and was classified as 1-, 2- or 3-vessel disease according to the presence of a stenosis greater than 50% of lumen diameter. Stenosis of the left main coronary artery of > 50% of lumen diameter was considered to be 2-vessel disease. Inter-observer variability of angiographic classifications was 4.9%.

Further quantitative angiographic evaluation was performed using an established scoring system [16]. Coronary segments were graded as grade 0, 1, 2, 3, 4 or occlusion based on the presence of < 25%, < 50%, < 75%, ≥ 75% or occlusion defined as a > 95% diameter stenosis with a severely reduced or no antegrade flow, respectively. CAD severity score was calculated as the average grade of the diseased coronary segments (i.e. ≥ grade 1). CAD extent score was calculated for each patient based on the number of segments exhibiting lesions ≥ grade 1.

The trial protocol specified IVUS examination of the same major epicardial coronary artery (preferentially the left-anterior descending coronary artery) and this was conducted while performing coronary angiography using a 20 MHz, 2.9F, monorail electronic Eagle Eye Gold IVUS catheter (Volcano Corporation Inc, CA, USA). IVUS images were acquired at a rate of 30 frames/s and pullback speed of 0.5 mm/s. Images were stored digitally for off-line analysis conducted after trial closure by a core laboratory (Oslo University Hospital, Rikshospitalet, Oslo, Norway) blinded to patient treatment. IVUS analysis was performed according to the guidelines for acquisition and analysis of IVUS images by the American College of Cardiology and European Society of Cardiology [17]. Contour detection of both the lumen and external elastic membrane (EEM) was performed at approximately 1 mm intervals using validated software (QIVUS, v.3.0, Medis medical imaging systems, Leiden, the Netherlands).

MIT was utilized as the primary grayscale IVUS efficacy variable. Previous studies have utilized MIT ≥ 0.5 mm as evidence of pathological intimal disease [12] and this cut-off was utilized in the current study. Other secondary IVUS variables were: (i) percent atheroma volume (PAV) which expresses the summation of atheroma areas in proportion to the EEM area using the equation: PAV = ∑ (EEM_area − Lumen_area)/∑EEM_area) × 100 and (ii) normalized total atheroma volume (TAV) using the equation: TAV = ∑ (EEM_area − Lumen_area)/number of frames) × median number of frames in cohort. IVUS endpoints in the ABCD sub-study population were compared with the non-T2D reference population of heart transplant donors.

Analyses were performed with the SPSS v 24.0 statistical software (SPSS Inc. Chicago, IL). Data is expressed as mean ± SD or as median (interquartile range) as appropriate and a two-tailed p-value < 0.05 was considered statistically significant. Baseline characteristics and IVUS endpoints were compared using Student’s t-test, Mann–Whitney test and Pearson’s Chi squared test as appropriate. Change in angiographic severity and extent score was compared between treatment groups by performing analysis of covariance (ANCOVA) with the baseline value included as a covariate and treatment group as a fixed factor. To account for an age-effect on atheroma-burden, IVUS data was also analyzed with age-stratification (< 50 years, 50–60 years and > 60 years).

When considering IVUS findings at 7 years according to baseline angiographic findings, there were no significant differences between the treatment groups in MIT or PAV regardless of whether baseline angiography had been normal (n = 21) or shown a stenosis of < 25% (n = 17) or > 25% (n = 18) (Additional file 2: Figure S2). Similarly, there was no significant difference in IVUS parameters between the treatment groups when patients were stratified according to baseline CAD extent score (Additional file 3: Figure S3).

CAD is the leading cause of morbidity and mortality in patients with T2D. Previously, T2D was considered a CAD equivalent by European and American guidelines [18] implying a high (> 20%) 10-year CV risk for all patients with T2D [18]. However, the validity of this assumption has been questioned in recent years based on data indicating the potentially wide heterogeneity in CV risk among T2D patients [19]. For example, a meta-analysis by Bulugahapitiya et al. included 45,108 patients and revealed a 43% lower risk of developing CAD in patients with T2D without prior myocardial infarction (MI) as compared to patients without T2D with previous MI [20]. Given this heterogeneity, various studies have been performed utilizing methods such as nuclear imaging, echocardiography, carotid ultrasound and exercise stress-testing, to evaluate the impact of non-invasive screening for CAD in asymptomatic patients with T2D [21]. According to these studies, the estimates of CAD prevalence in asymptomatic patients with T2D vary widely with results ranging from 8 to 50% [22]. For example, the DIAD study [8] showed that 22% of patients had abnormal stress myocardial perfusion imaging whereas Rajagopalan et al. [23] found that 58% of asymptomatic patients with T2D patients had abnormal SPECT imaging.

Importantly, surrogate markers of CAD obtained from non-invasive testing carry important prognostic information in asymptomatic individuals with T2D, including coronary artery calcium (CAC) by coronary computed tomography angiography (CTA) [24, 25], carotid atherosclerosis [26] and plaques [25], and aortic stiffness [27]. Despite this and the diverging results of occult CAD prevalence, current guidelines do not recommend universal screening for CAD among patients with T2D, partly based on coronary computed tomography angiography (CTA) studies [28, 29] showing that a sizeable proportion of patients (25–30%) do not have demonstrable plaque on CTA [19]. The ABCD trial is the first trial to date to perform invasive IVUS imaging, a gold-standard technique for evaluation of CAD, in asymptomatic patients with T2D and reveals a CAD prevalence of 84%, indicating a significantly higher burden of disease than previously assumed. Given the increased vulnerability observed in coronary plaques in subjects with diabetes and CAD [30] this prevalence is important to consider. Current guidelines although differentiating their recommendations for choice of glucose lowering medication according to the presence or absence of cardiovascular disease, do however not consider prevalent occult CAD in T2D patients and a refined accurate estimate of risk and disease prevalence is warranted in an effort to improve the applicability and validity of T2D management guidelines.

A study from 1984 found no significant progression in extent score in a population with CAD of which the majority had their second angiogram performed due to persistent angina [31], but they found that high extent score was an independent, strong predictor of CAD progression. Despite the relatively low extent score in our study, and the high use of statins, we found a substantial progression in both the extent and severity of CAD. This confirms a more aggressive atherosclerosis seen in T2D, and is in line with a Korean study that demonstrated increased progression of coronary artery calcification in those with diabetes as compared to those without [32].

A previous study in patients with T2D and microalbuminuria from the pre-statin era, demonstrated that an intensive multi-interventional therapy program aimed at behavioral modification and pharmacologic therapy targeting hyperglycemia, hypertension, dyslipidemia, and microalbuminuria reduced the risk of CV and microvascular events by about 50 percent [33]. The ABCD trial has previously reported that the 2 year structured, hospital based multi-intervention significantly reduced estimated CV risk in T2D patients [13], however, a subsequent long-term follow-up failed to demonstrate an improvement in CV outcome and mortality [15]. This is congruent with a recent study by Ueki et al. of the Japan Diabetes Outcome Intervention Trial 3 (J-DOIT3) [34], where also a lack of benefit on mortality and CV events was reported despite 8.5 years of effective multi-factorial, target-driven treatment in patients with T2D with additional CV risk factors. The current report of the ABCD trial supports these findings as evidenced by no significant difference in progression of angiographic CAD between the multi-interventional and standard group. Furthermore, there was no significant difference between the multi-interventional and standard group in IVUS parameters 7 years after randomization. However, interpretation of the effect of multi-interventional therapy on CAD assessed by IVUS is limited by the lack of baseline IVUS imaging that does not allow evaluation of disease progression from baseline.

The current neutral results of multi-interventional treatment on CV outcome and CAD extent/severity progression are in concordance with two previous landmark trials. The ACCORD study [35] reported that the use of intensive therapy targeting HbA1c levels did not significantly reduce major cardiovascular events or mortality. The Factor 64 randomized trial [36] utilized coronary CTA to identify CAD in patients with diabetes, and although more aggressive medical therapy in those identified with CAD had a positive effect on lipids, blood pressure, and glucose control, there was no impact on death and coronary heart disease outcomes. Our results conflict however with the DIANA study [37], which showed decreased CAD progression rate with improved glycaemic control after 1 year treatment with voglibose or nateglinide in early T2D. Nevertheless, an intervention period longer than the 2 years in the ABCD trial may ultimately be required to demonstrate a beneficial effect of a multi-interventional strategy on CV outcome and CAD in this population with more advanced T2D and further research is warranted. Furthermore, a more individualized multi-interventional treatment strategy with incorporation of newer therapeutic agents, such as, sodium glucose co-transporter-2 (SGLT-2) inhibitors [38], or glucagon like peptide-1 (GLP-1) agonists [39] with evidence for benefit in patients with CV disease manifestations, may be required to reduce the residual CV risk.

The present study has some limitations. The intervention period was relatively short and the number of patients was relatively small, with also an unintended observation of small baseline imbalances in CAD extent and severity between the treatment groups, which may have influenced the chance to modulate the progression by the intervention. IVUS imaging was not performed at baseline and this limits the possibility for an accurate assessment of CAD disease progression over 7 years in the two treatment groups. Furthermore, the reference population of heart transplant donors is a selected population free from CV risk factors and of notably lower age than the T2D cohort, and may not be representative of the general population. However, the authors believe the data is unique as it provides a gold-standard assessment of CAD prevalence in asymptomatic patients with T2D that challenges the assumptions of current guidelines that are based on non-invasive and less sensitive methods to diagnose CAD.