Six-Year Outcomes of CABG vs PCI in Diabetic Patients with Multivessel Coronary Disease

Introduction

Despite the efforts to improve prevention and treatment, coronary artery disease (CAD) is still among the leading causes of death and disability, with an estimated prevalence of 244.1 million cases worldwide in 2021 [1,2]. One of the main drivers of coronary vascular damage is diabetes. The global prevalence of diabetes among adults aged 20–79 in 2021 was estimated to be 10.5% and is expected to rise to 12.2% by 2045 [3]. This is due to the increasing prevalence of diabetes risk factors, such as obesity, sedentary lifestyle, and poor diet [4]. Diabetes can be diagnosed based on: a) hemoglobin A1C ≥6.5%, b) fasting plasma glucose ≥126 md/dl, c) 2-hour plasma glucose ≥200 mg/dl in oral glucose tolerance test, or d) a random plasma glucose ≥200 mg/dl in the presence of symptoms [5]. The presence of diabetes significantly decreases survival in patients with multivessel disease (MVD) compared to those without diabetes [6].

Because of the high comorbidity burden and the disease complexity among patients with MVD, multidisciplinary Heart Team (HT) evaluations are widely incorporated in the decision-making process. Institutional decisions made by a team consisting of an invasive cardiologist, cardiac surgeon, clinical cardiologist, and, if needed, other specialists are reproductible and less prone to bias and therefore have earned a class I recommendations in European and American revascularization guidelines [7–9].

According to current guidelines, CABG is preferred over PCI in the case of MVD with left main (LM) involvement or diabetes mellitus, especially in the presence of complex lesions [7,8,10]. In patients with diabetes, CABG has been consistently superior to PCI in repeat revascularization (RR) endpoint, but most of the randomized studies were conducted before the use of modern drug-eluting stents [10–14]. Some studies suggest that the progress in percutaneous techniques has narrowed this gap and there no longer is a mortality benefit in LM lesions, although higher disease complexity corresponds with greater CABG advantage [15]. The mortality benefit of CABG in patients with diabetes has also been shown in multiple studies [15,16]. Nevertheless, the difference in mortality might be more closely related to the completeness of revascularization than the chosen treatment method, as incomplete revascularization was found to increase overall mortality in both CABG and PCI groups [17,18]. While diabetes is a known risk factor for multivessel coronary artery disease, there is a paucity of data regarding the impact of Heart Team decision-making on the outcomes of revascularization in diabetic MVD patients. In this post hoc analysis of a larger, previously described cohort, we review the Heart Team decisions and treatment patterns of diabetic MVD patients treated in a single tertiary-care center, focusing on the differences between CABG and PCI cohorts.

Material and Methods

STATISTICAL ANALYSIS:

The PQStat software (version 1.6.6, PQStat, Poznań, Poland) was used for statistical analysis. The normality of distribution for continuous variables was confirmed with the Shapiro-Wilk test. Categorical data are expressed as counts and percentages, while continuous data were presented as means with standard deviation (SD). Comparisons between groups were performed using the Pearson’s chi-squared test for categorical variables, and either the t test or Mann-Whitney U test for continuous variables, according to data distribution. Comparisons between groups of patients qualified for individual treatment strategies were performed using the chi-square test, and statistical analysis used one-way analysis of variance (ANOVA). Time-to-event analysis was performed using Kaplan-Meier curves. All P values (P) are 2-sided, and P values lower than 0.05 were considered statistically significant.

Results

STUDY POPULATION:

Between January 2016 and December 2019, 1035 patients with MVD underwent HT evaluation. After applying the inclusion and exclusion criteria, total of 317 individuals with diabetes remained. The mean (SD) follow-up time was 72 (26) months. The mean (SD) age of overall cohort was 69.3 (9.4) years, 67.5% were men, the mean (SD) body mass index (BMI) was 27.8 (3.8) kg/m2, and 11.7% of patients had frailty syndrome. The surgical risk was calculated using 2 scores – the mean (SD) overall values of EuroSCORE II (European System for Cardiac Operative Risk Evaluation II) and the mean STS (Society of Thoracic Surgeons) scores were 6.5 (3.4)% and 4.1 (1.9)%, respectively. The 2 study groups differed significantly in rates of frailty, active cancer, EuroSCORE II score, STS score, and a rate of complete revascularization. Detailed data are presented in Table 1. Among all enrolled patients, common CAD risk factors were relatively high – 83.6% had hypertension, 82.3% had dyslipidemia, and 26.2% were current smokers. One-third of patients required treatment with insulin. Patients qualified for PCI were more burdened and presented more often, but the difference between the 2 groups was not statistically significant – chronic obstructive pulmonary disease (COPD) was 18.3% vs 10.2%, P=0.07, severe PH was 12.8% vs 6.1%, P=0.08, CKD was 45.7% vs 36.7%, P=0.14, atrial fibrillation (AF) was 29.7% vs 23.4%, P=0.26, anemia was 40.2% vs 29.6%, P=0.07, peripheral artery disease (PAD) was 14.2% vs 8.2%, P=0.13, and history of prior stroke/transient ischemic attack (TIA) was 15.1% vs 9.2%, P=0.15. Furthermore, 48.9% of participants presented with acute coronary syndromes (ACS), while the rest had chronic symptoms of CAD. The rates of ST-segment elevation myocardial infarction (STEMI) and non-STEMI (NSTEMI) were insignificantly higher in the PCI group (5.5% vs 2.0%, P=0.17 and 45.7% vs 41.8%, P=0.53, respectively). The prevalence of stable CAD was higher in the CABG group (56.1% vs 48.9% for PCI; P=0.23). The rates of prior myocardial infarction (MI) and prior revascularization were similar between groups (39.2% vs 28.6% for CABG, P=0.08, and 43.4% vs 32.7% for CABG, P=0.07, respectively). The prevalence of congestive HF was relatively low overall (36.0%), but was insignificantly higher in the PCI group (38.8% vs 29.6% for CABG, P=0.11). The PCI group had lower (mean [SD]) left ventricle ejection fraction (LVEF) (44.3 [6.9]% vs 46.9 [7.1]%, P<0.37). Baseline clinical characteristics (overall and by groups) in details are presented in Table 1.

ANGIOGRAPHIC PARAMETERS:

Based on angiographic characteristics, the burden and complexity of CAD were high overall. The mean (SD) number of affected lesions was 4.3 (1.6). Significant LMD was present in 35% of all patients, coronary stenosis involving bifurcation was found in 70.1%, one-third of all patients had severe calcification, and 28.1% was diagnosed with at least 1 chronic total occlusion (CTO). The mean (SD) numbers of stents implanted was 2.6 (1.3), and 2.9 (1.5) conduits (arterial and venous) were performed per patient. The mean (SD) SYNTAX score overall was 30.9 (6.5) points – 31.1 (6.0) in the CABG group and 30.9 (6.7) in the PCI group (P=0.79). See Table 2.

MEDICATIONS ON ADMISSION AND AT DISCHARGE:

Adherence to guideline-directed medical therapy was high, although drugs use differed between the groups during follow-up. On admission, there was no difference in use of ACE (angiotensin-converting enzyme)-inhibitors, ARBs (angiotensin receptor blockers), statins, and beta-blockers (BB). At discharge, the PCI group had a higher percentage of individuals taking aspirin, P2Y12 inhibitors, loop diuretics, and aldosterone antagonists, while loop diuretics were more frequently prescribed in the CABG group. The use of metformin and SGLT-2 (sodium-glucose cotransporter-2) inhibitors was comparable between groups. See Table 3.

OUTCOMES:

Six-year follow-up was achieved in 100% of patients. The incidence of all-cause death overall was 19.2%, with no significant differences between the CABG (16/98 [16.3%]) and PCI (45/219 [20.5%]) groups, P=0.38 (Figure 2). The PCI group had a higher rate of MACCE, mainly driven by higher rates of RR – 183/219 (83.6%) vs 44/98 (44.9%) for CABG, P<0.01, and 103/219 (47.0%) vs 17/98 (17.3%) for CABG, P<0.01, respectively). The PCI group had a higher rate of MI – 27/219 (12.3%) vs 4/98 (4.1%) for CABG, P=0.02). The CABG group had significantly longer postprocedural hospital stay – mean (SD): 10.7 (1.9) vs 4.4 (0.8) days for PCI, P<0.01. Rates of stroke and in-hospital mortality were comparable between the 2 groups – 8.2% vs 5.9% for PCI, P=0.46 and 2.0% vs 0.5% for PCI, P=0.18, respectively (Table 4).

Discussion

LIMITATIONS:

Due to the retrospective nature of our study, the risk of bias could be greater than with RCTs. Moreover, because of its design, this study may lack the statistical strength to determine differences in mortality. Because this was a single-tertiary-center study, the sample size was limited and patient group characteristics might differ from a broader population. Due to site-specific conditions, the results might not be fully reproducible in other centers. Furthermore, there were minor differences between baseline characteristics of the 2 study groups. The HT evaluations may be team-specific and do not reflect decisions made in other hospitals. The experience of surgical and percutaneous teams cannot be overlooked. Patient medication compliance could not be objectively measured. Moreover, a hierarchy within HT also exists, which could have led to bias.

Conclusions

Heart Team qualification for PCI is more commonly implemented in more burdened, frail patients, with a higher surgical risk. Overall mortality did not significantly differ between the PCI and CABG groups, but some disparities in secondary endpoints, revealed in previous RCTs, were observed. The choice of revascularization strategy for heavily burdened patients should never be individual and Heart Team involvement in decision making is essential.