CABG: Coronary Artery Bypass Grafting; CAD: Coronary Artery Disease; CI: Confidence Intervals; CKD: Coronary Kidney Disease; HR: Hazard Ratio; PCI: Percutaneous Coronary Intervention; eGFR: Estimated Glomerular Filtration Rate; CKD-EPI: Chronic Kidney Disease Epidemiology Collaboration

Chronic kidney disease (CKD) has numerous direct negative effects on the cardiovascular system [1-3]. The effects of decreased renal function on cardiovascular disease are attributed to coexisting classical cardiovascular risk factors (e.g., diabetes, hypertension, smoking, dyslipidemia, and advanced age) and non-classical CKD-related risk factors (e.g., mineral and bone disease abnormalities, anemia, inflammation, and oxidative stress) [4]. A subclinical atherothrombotic process based on oxidative stress, inflammation, and endothelial dysfunction has been observed to exist in the early stages of renal dysfunction [5]. These metabolic changes progress with renal impairment, eventually leading to more advanced atherogenic burden at the systemic level [6]. For risk stratification in patients with multivessel coronary artery disease (CAD), understanding the relative effect of coronary revascularization according to the stage of renal dysfunction is crucial. However, the differences in the clinical features and outcomes of patients undergoing coronary revascularization according to the varying degree of renal dysfunction remain unknown. Moreover, as patients with mildly decreased renal function corresponding to the borderline between the CKD and normal renal function stage were under-represented in most of the clinical trials on myocardial revascularization, there are insufficient data on the characteristics and long-term prognosis of patients with mildly decreased renal function after percutaneous or surgical revascularization [7-9]. Resultingly, the level of renal function at which cardiovascular risks begin to increase has not been clearly elucidated.

By reviewing a recent published manuscript titled, “Prognostic Impact of Mildly Impaired Renal Function in Patients Undergoing Multivessel Coronary Revascularization” [10], we summarized the prognostic impact of baseline renal function in patients who underwent multivessel coronary revascularization, focusing more on patients with mild renal dysfunction.

This study was a registry-based analysis on consecutive patients with multivessel CAD who underwent either isolated coronary artery bypass graft (CABG) surgery or percutaneous coronary intervention (PCI) with drug-eluting stents between January 1, 2003 and April 30, 2018, at Asan Medical Center (Seoul, Korea) [11]. A total of 10,883 eligible patients with multivessel disease who underwent coronary revascularization were identified, of whom 6,730 (61.8%) patients underwent PCI with drug-eluting stents and 4,153 (38.2%) patients underwent isolated CABG.

Given that serum creatinine is considered as an insensitive indicator of renal function and mild renal impairment exists despite normal or near-normal creatinine levels [12], the estimated glomerular filtration rate (eGFR) was used for the assessment of renal dysfunction in the current study. The eGFR at baseline was measured using the CKD Epidemiology Collaboration (CKD-EPI) creatinine equation that includes age, race, and sex as surrogates for the non-GFR determinants of serum creatinine [13]. According to the baseline eGFR, 3,735 (34.3%) patients were considered to have normal renal function (stage I), while 5,125 (47.1%), 1,497 (13.8%), 198 (1.8%), and 331 (3.0%) patients were deemed to have mild (stage II), moderate (stage III), severe (stage IV), and end-stage renal dysfunction (stage V), respectively. The primary outcome was the composite of death, spontaneous myocardial infarction, or stroke. All the outcomes were assessed according to standard endpoint definitions [14].

In general, patients with more severe renal dysfunction showed a higher prevalence of clinical risk factors (such as old age and left ventricular dysfunction) and comorbidities (such as diabetes, hypertension, previous stroke, and previous myocardial infarct). The prevalence of three-vessel disease and left main disease was also higher in patients with more severe renal dysfunction. The difference in baseline demographics according to the stage of renal function was more prominent in the advanced renal dysfunction group (stage IV/V). In the overall population, the risk of the primary composite outcome of death, spontaneous MI, or stroke proportionally increased, especially in patients with more severe renal dysfunction (stage IV and V) with respect to the predicted probability per 1,000 person-year of events (Figure 1A). This increased risk of primary composite outcome associated with decreased renal function was not different in each stratum of PCI or CABG (Figure 1B and 1C).

The multivariable-adjusted hazard ratios (HRs) for outcomes according to the stage of renal function in the overall population are shown in Figure 2A and those in each stratum of PCI and CABG are shown in Figure 2B, 2C. With the stage I group as the reference, the adjusted HRs for primary composite outcome proportionally increased according to increasing levels of renal dysfunction (HRs for stage II, stage III, stage IV, and stage V: 1.16, 1.74, 3.70, and 4.30, respectively; P-for-trend <0.001). This pattern was consistent in each stratum of the PCI and CABG groups.

Using the linear model, we calculated the 10-year probability of the primary composite outcome according to the baseline eGFR (as a continuous variable). In the crude model, the probability of primary composite outcome substantially increased below the eGFR value of approximately 60 ml/min/1.73 m². In multivariable models, although the effect of eGFR on outcome was slightly diminished, the inverse relationship of eGFR with the risk of 10-year probability of the primary outcome was consistent (Figure 3A). This association was similar in each stratum of PCI and CABG (Figure 3B).

Although the severity was less than patients with advanced renal dysfunction, patients with mildly decreased renal function were also older and had a relatively higher prevalence of comorbidities and more extensive CAD compared to patients with normal renal function. To rigorously control the baseline characteristics of patients according to each stage of renal function and different revascularization strategies of PCI and CABG, propensity-score matching was used to assemble a cohort of patients with similar baseline characteristics. In the propensity-matched cohort, the adjusted risk for primary composite outcome was not significantly different between the stage I and stage II groups (HR: 1.12; 95% CI: 0.97–1.30; P = 0.12). However, the risk of primary composite outcome was significantly higher in the stage III group than in the stage I group (HR: 1.50; 95% CI: 1.22–1.84) (Table 1 and Figure 4). The relative effect of PCI over CABG for the primary outcome was similar in the matched cohort of each renal function group (HR: 1.17; 95% CI: 0.89–1.54 for stage I, HR: 0.94; 95% CI: 0.78–1.13 for stage II, and HR: 1.02; 95% CI: 0.80–1.30 for stage III) (Table 2 and Figure 4).

In the present study, we included a large number of patients with a broad spectrum of renal function, which enabled us to perform a detailed evaluation of the clinical impact of baseline renal function after coronary revascularization. Overall, we observed substantial differences in the demographic features, comorbidities, clinical presentation, and atherosclerotic burden of CAD according to the degree of renal dysfunction. When focusing on the early stages of renal dysfunction, as renal function deteriorated from mild to moderate dysfunction, the patients were older, with more frequent comorbidities, and more advanced coronary disease. After an adjustment of baseline covariates using a propensity-score analysis, compared to patients with normal renal function, those with mild renal dysfunction did not have increased risks of the primary composite outcome of death, spontaneous MI, or stroke. By contrast, patients with moderate renal dysfunction had a higher risk of major adverse cardiovascular events.

These findings may define a threshold of GFR that is associated with worse outcomes and suggest that mildly decreased renal function is not a critical prognostic parameter after coronary revascularization. Considering the mechanisms underlying the poor outcomes of renal impairment following coronary revascularization are likely multi-factorial, such findings may be explained in part by that the detrimental synergism between coexisting classical cardiovascular risk factors and non-classical CKD-related risk factors is not yet prominent in patients with mildly decreased renal function. It was observed that the atherothrombotic process was actually subclinical in the stage of mild renal dysfunction, whereas in CKD, the systemic atherogenic process based on oxidative stress and inflammation was more enhanced with the coexisting traditional cardiovascular risk factors, resulting in a higher risk of adverse clinical events after coronary revascularization (Figure 5) [15].

As PCI might be more challenging with a lower probability of procedural success and a higher proportion of incomplete revascularization due to more aggressive atherosclerosis and a higher degree of calcification in patients with advanced renal disease, it is generally accepted that CABG is superior to PCI in achieving better long-term outcomes in the highest-risk CKD patients [16]. However, since most previous clinical trials did not distinguish patients with mildly decreased renal function from those with normal renal function or advanced CKD, the comparative outcome of two revascularization strategies in such borderline-risk patients remains uncertain. In our large-sized real-world study, there was no significant difference between PCI and CABG with respect to the risks of the primary composite endpoint in each group of normal, mild, or moderate renal dysfunction. Well-organized understanding of the clinical features and prognostic impact of mildly decreased renal function, and the relative effect of PCI or CABG in patients with mildly decreased renal function may be helpful for further risk-stratification and crucial to guide for appropriate revascularization choices. However, owing to the inherent limitations of an observational study, this interpretation should be considered in a conservative manner; our observed findings warrants further investigation and should be confirmed or refuted through large, randomized clinical trials with a long-term follow-up.