Renal Dysfunction Increases Risk of Adverse Cardiovascular Events in 5-Year Follow-Up Study of Intermediate Coronary Artery Lesions

Introduction

Progression of chronic coronary syndrome (CCS), namely ongoing atherosclerotic plaque formation and risk of plaque destabilization, depends on various factors, including comorbidities [1–3]. Chronic kidney disease (CKD) is associated with a worse prognosis in patients with CCS [1,4,5]. Several studies have shown the association between deteriorating renal function and an increased risk of acute coronary syndrome (ACS), as well as a higher prevalence of major adverse cardiovascular events (MACE) after percutaneous coronary intervention (PCI) [1,4,5]. Moreover, atherosclerosis has been proven to develop more rapidly in patients with CKD, especially those with end-stage kidney disease [6,7]. Optical coherence tomography (OCT), fractional flow reserve (FFR), and intravascular ultrasound (IVUS) of intermediate-grade coronary stenoses (40–80%) [8,9] can provide meticulous information regarding plaque composition and influence on hemodynamic function [10–13]. In severe CKD patients, a larger amount of calcium is observed and the calcified plaque is more prevalent [14–19], demanding a different approach during PCI, including the use of high-pressure balloons, cutting balloons, intravascular lithotripsy, or rotablation [20]. CKD can also lead to higher FFR values [21]. There is limited information on specific, multimodality-derived plaque characteristics in patients with mild-to-moderate renal dysfunction and the association with follow-up.

Due to the widely recognized limitations of coronary angiography in determining hemodynamic significance of ambiguous coronary stenoses, functional lesion assessment is recommended [22–24] along with intravascular imaging, and this had gained increasing interest in the management of coronary syndromes [25–32]. FFR, a wire-based technique, is performed during cardiac catheterization, measuring the pressure before and after the stenosis in an isolated segment of the vessel during maximal hyperemia [33], which is usually induced by intravenous administration of adenosine [34]. The result is presented as a ratio of pressure distal to the stenosis divided by pressure proximal to the stenosis and is a useful tool for identifying intermediate-grade lesions significantly restricting blood flow [35]. OCT has some distinctive features, including high resolution (axial 10–20 μm and lateral 20–90 μm) and penetration depth ranging from 1 to 2 mm [25]. It thus allows a thorough analysis of the superficial vessel layers, and a detailed coronary plaque assessment is compromised by imperfect visualization of the deeper layers of the coronary artery [25,36–42]. Intravascular ultrasound (IVUS) has a tissue penetration depth of 5–6 mm, allowing visualization of the full vessel wall thickness, including the external elastic membrane (EEM), the diameter of which serves as a guide for stenting optimization, but with a lower resolution (axial 100–150 μm) than OCT [26,36,37,42–44]. Additionally, the ability to visualize all layers of the arterial wall enables assessment of vessel remodeling and plaque burden [37,45].

Research is ongoing regarding high-risk plaque characteristics, which in the long term are likely to cause MACE [28,46–53]. Assessing coronary arteries multimodally allows for acquisition of data regarding plaque morphology and functional indices [28,47,50–53]. Patients with CKD are more susceptible to faster progression of CCS [1,4,5]. There is limited information on the impact of mild-to-moderate renal dysfunction on intermediate-grade stenoses in coronary arteries in relation to long-term follow-up. The aim of this 5-year follow-up study was to assess the morphometric (OCT, IVUS) and functional (FFR) parameters of intermediate coronary artery lesions evaluated during cardiac catheterization in patients with renal dysfunction vs those without renal dysfunction.

Material and Methods

ETHICS APPROVAL AND INFORMED CONSENT:

Approval for this prospective study was granted by the Ethics Committee at the Medical University of Warsaw (approval no. 1432020), and all procedures were conducted in adherence to the principles outlined in the Declaration of Helsinki. Prior to participation, patients provided informed and written consent.

STUDY POPULATION:

This was a single-center, prospective, observational, longitudinal cohort study, which consecutively enrolled 100 patients (118 lesions) with CCS undergoing a coronary angiogram (CAG) in a tertiary heart center (ClinicalTrials.gov registration nr: NCT06261866) [54,55]. Intermediate-grade coronary lesions were analyzed by a multimodality approach using OCT and FFR, and with IVUS in a subset of patients. Stenoses were classified as significant if FFR was ≤0.8. Patients were divided into 2 groups according to the estimated glomerular filtration rate (eGFR) level: 45–60 ml/min/1.73 m2 for mild-to-moderate renal dysfunction and >60 ml/min/1.73 m2 for no renal dysfunction.

The inclusion criteria were: CCS defined as a positive result of an ischemic test (exercise test or single photon emission tomography (SPECT)) or prevalence and severity of chest pain symptoms ranked according to the Canadian Cardiovascular Society classification (CCS 2–3), age >18 years, and intermediate coronary stenoses defined as a stenosis of 40–80% evaluated visually in CAG [9]. Exclusion criteria were: hemodynamic instability, bypass graft or ostial left main or ostial right coronary lesions, severely impaired renal function defined as eGFR <45 ml/min/1.73 m2, contraindication for adenosine administration, and pregnancy.

OPTICAL COHERENCE TOMOGRAPHY:

OCT recordings and data acquisition were performed utilizing an OCT imaging system (Abbott, C7XR Dragonfly TM, LightLab Imaging, Inc., MA, USA). Images were obtained using the non-occlusive technique [22,25,41,42,56,57]. This technique does not require the use of a balloon to block the coronary blood flow for the image acquisition; instead, a quick flush of contrast is administered into the coronary artery, thus blood morphological elements do not cause artifacts [56,57].

Current expert consensuses [58–61] served as guidelines for OCT images analysis. Analysts were kept unaware of FFR, IVUS results, and patients’ characteristics. Evaluation of the reference lumen area involved assessment of the largest lumen either proximal or distal to a stenosis (within 10 mm of the stenosis). Morphometric assessment of plaques was conducted at the site of minimal lumen area (MLA) over at least 3 consecutive frames. Plaques were categorized as calcified, fibrous, lipid-rich, or mixed. Calcified plaques exhibit a signal-poor heterogeneous region with sharply delineated border [58–61]. We also assessed the calcium angle within the plaque. A fibrous plaque is defined by high backscattering and a relatively homogenous signal [58–61]. A plaque was deemed lipid-rich when it presented an irregular signal-poor region with diffused borders (Figure 1) [59,62]. The lipid angle was calculated as the degree measure of a low-signal region. Fibrous cap thickness (FCT) refers to the distance between the inner edge of the lipid or calcium pool and vessel lumen border. FCT measurements were taken at 0.2-mm intervals across the plaque. Subsequently, 3 measurements were taken at the thinnest part of the plaque at each cross-section, and the average value was considered for the final analyses [62]. A plaque with minimal FCT <65 μm was classified as a thin-cap fibroatheroma (TCFA) [61].

FRACTIONAL FLOW RESERVE:

The coronary pressure analysis was performed with a commercially available 0.014-inch pressure guide wire (St. Jude Medical, Minneapolis, MN, USA). The pressure wire was placed into the coronary artery, then the tip of the pressure wire was accurately positioned 1–2 mm outside of the catheter to equalize pressure [63]. Afterwards, the pressure wire was placed distally to the assessed lesion [63] and maximal hyperemia was induced by administering adenosine intravenously at a rate of 140 μg/kg/min through a large peripheral vein [34]. FFR was determined by dividing the mean hyperemic distal coronary pressure by the mean aortic pressure. A lesion was considered significant if the FFR was ≤0.8 [12,13,64–66].

INTRAVASCULAR ULTRASOUND:

For IVUS image acquisition, a catheter was positioned in the distal section of the coronary vessel and a pullback was conducted at a speed of 0.5 mm/s at 40 MHz [67]. Image evaluation occurred at 0.5-mm intervals utilizing specialized software, with the analyst being blinded to patients’ FFR and OCT results, as well as clinical characteristics. The plaque burden at the site of MLA was computed using the formula: [68].

FOLLOW-UP:

The primary endpoint was MACE defined as a composite of cardiac death, myocardial infarction, stroke, unplanned PCI, coronary artery bypass grafting and hospitalization due to heart failure. Cardiac death was defined as death due to CCS, ACS, or cardiac arrest. The secondary endpoint was all-cause death. The follow-up was obtained by on-site visits, through data obtained in national and hospital registries, and by phone if necessary.

STATISTICAL ANALYSIS:

Statistical analyses were conducted using SPSS version 28.0 (IBM Corp, Armonk, NY, USA). The distribution of collected data was assessed through the Kolmogorov-Smirnov test. For normally distributed data, means and standard deviations (SD) were presented, while non-parametric data were represented by median values and the interquartile range (IQR) between the 25^th and 75^th percentiles. Categorical data were expressed as percentages within respective groups. Group comparisons involved the t test for normally distributed data and the Mann-Whitney U test for non-normally distributed continuous variables. Categorical variables were analyzed using the chi-square test with Yates’s correction for continuity when applicable. Statistical significance was determined at a double-sided P value of less than 0.05. Survival and MACE-free survival follow-up were assessed using the log-rank test and illustrated through Kaplan-Meier curves.

Results

PATIENTS’ CHARACTERISTICS:

This study enrolled 100 patients (118 lesions) with a mean age of 64.5±9.5 years. Patients were divided into 2 groups – renal dysfunction (n=23) and no renal dysfunction (n=77) – according to the eGFR level: 45–60 ml/min/1.73 m² for mild-to-moderate renal dysfunction and >60 ml/min/1.73 m² for no renal dysfunction. The median value of eGFR in both groups was 54.9±3.9 ml/min/1.73 m² vs 85.8±17.1 ml/min/1.73 m², for renal dysfunction and no renal dysfunction, respectively (P<0.001).

The baseline characteristics are presented in Table 1. Patients with renal dysfunction were older (67.7±8.1 vs 63.6±9.7 years, P=0.044). No significant dissimilarities were found regarding comorbidities or laboratory test results (except creatinine level 1.25±0.16 mg/dl vs 0.89±0.16 mg/dl, renal dysfunction vs no renal dysfunction respectively, P<0.001) between groups. Patients without renal dysfunction were more frequently treated with β-blockers (97.4% vs 87.0%, P=0.044) (Table 1).

FRACTIONAL FLOW RESERVE:

All lesions were assessed by FFR. The plaque location did not manifest significant dissimilarities, and in both groups most frequently appeared in the left anterior descending artery. There was a trend in difference in FRR values between renal dysfunction patients and no renal dysfunction patients (0.81±0.11 vs 0.77±0.12, respectively, P=0.069).

OPTICAL COHERENCE TOMOGRAPHY:

In OCT analysis, no differences were shown regarding plaque type in both groups. Among renal dysfunction patients, the 2 most prevalent plaque types were fibrous and calcified (both: n=11, 36.7%). On the contrary, the most common plaque type in patients without renal dysfunction was calcified (n=32, 37.2%) (Table 2).

There were no dissimilarities regarding lesion length, minimal lumen diameter, minimal lumen area, mean lumen area, cap thickness over calcium, angle of calcium, presence of TCFA, and thrombus (Table 3).

INTRAVASCULAR ULTRASOUND:

In a subset of 57 lesions, IVUS was performed. As with OCT, there was no statistically significant difference in lesion length, lumen volume, plaque volume, lumen area, vessel area, or plaque burden between groups (Table 4).

FOLLOW-UP:

Clinical follow-up data on mortality were available for 95 study participants. The median follow-up time was 55.5±12.5 months. Presence of mild-to-moderate renal dysfunction in 5-year long follow-up was linked to an increased risk of mortality from any cause (33.3% vs 12.2%, HR=3.16, 95%CI [1.18, 8.51], P =0.016) in multivariate analyses adjusted for age and sex (Figure 2).

Follow-up regarding MACE (defined as cardiac death, myocardial infarction, stroke, unplanned PCI, coronary artery bypass grafting, or hospitalization due to heart failure) was available for 62 patients (Table 5). The median follow-up time was 45.8±20.3 months. MACE in 5-year follow-up occurred in 70.6% of patients with mild-to-moderate renal dysfunction compared to 36.6% of patients without renal dysfunction (HR=2.54, 95% CI: 1.20–5.38, P=0.011 in a multivariate analysis adjusted for age and sex) (Figure 3).

Discussion

The main findings of this study can be summarized as follows: i) Patients suffering from CCS with mild-to-moderate renal dysfunction did not present significant differences in OCT-detected plaque morphology and IVUS-obtained parameters compared to individuals without renal dysfunction; ii) There was a trend suggesting that renal dysfunction patients tend to have overall higher FFR values; iii) Renal dysfunction was linked to elevated risk of all-cause mortality and increased risk of MACE in 5-year follow-up.

A study by Tebaldi et al that enrolled 1004 patients, including 131 with eGFR values of ≤45 ml/min/1.73 m2, showed that patients with CKD present significantly higher FFR values in comparison to people with mild-to-moderate renal dysfunction or no renal dysfunction (0.84±0.07 vs 0.81±0.08, P<0.001) [21]. Creatinine clearance ≤45 ml/min was an independent predictor of FFR measurement (≤0.8 vs >0.8) [21]. The trend in FFR values noted in our investigation, as well as the results of the study by Tebaldi et al, suggest a negative correlation between eGFR and FFR values. The primary factor influencing FFR is the stenosis severity, but it also depends on the ability of the vessel to dilate after adenosine administration. It has been previously shown that CKD patients are characterized by endothelial and microvascular dysfunction leading to a reduced coronary vasodilator capacity [69,70]. It is possible that this microvascular dysfunction can reduce hyperemia after adenosine administration, resulting in a higher rate of negative FFR values.

In our IVUS analysis, no differences were established regarding: lumen volume, plaque volume, lumen area, plaque area, plaque burden, and calcification angle. A study by Miyagi et al divided 89 patients with stable angina into 2 groups in relation to eGFR with a cut-off point at 60 ml/min/1.73 m2, and they had a negative correlation with lipid volume [18]. Two studies have concentrated on IVUS evaluation of coronary plaques in hemodialyzed patients and showed more calcifications and a greater plaque volume in this group [17,19]. The primary difference between those studies in relation to our study is the enrollment criteria. Hemodialyzed patients are an extreme example of chronic kidney disease with minimal eGFR values; in contrast, the mean eGFR value in the renal dysfunction group in our study was 54.9±3.9 ml/min/1.73 m2.

We found no statistically significant differences regarding OCT-defined plaque types in renal dysfunction vs no renal dysfunction patients. Several studies have found more frequent calcifications among patients with CKD [14–16]. Of note, however, those studies concentrated on patients with end-stage renal disease and hemodialyzed patients, whereas our study concentrated on people with mild-to-moderate renal dysfunction. As such associations were not found in our study, it is crucial to further deepen the research regarding the number of calcifications among people with CKD, stratifying the participants into subgroups based on eGFR levels. A study be Sugiyama et al was designed to compare OCT-derived plaque parameters in different CKD stages. In patients presenting with CCS, 296 native coronary lesions treated with stent implantation were assessed. Lesions were divided into 3 groups: a non-CKD group (eGFR≥60 ml/min/1.73 m2), a CKD group (eGFR 15–60 ml/min/1.73 m2), and an end-stage kidney disease (ESKD) group (eGFR<15 ml/min/1.73 m2). Lipid-rich plaques were more frequent in the CKD group (P=0.01) compared to the non-CKD group, but no differences were found in calcification arcs, whereas when compared to the ESKD group calcifications were less prevalent (P=0.025) [71]. Okamura et al came to the conclusion that the amount of calcium in OCT was positively correlated with the duration of dialysis and consequently the duration of CKD [16], and the deterioration of renal function and advancement into the ESKD category was positively correlated with the progression of calcifications [71]. There were also contrasting results in terms of lipid arc. The study by Chin et al demonstrated no differences between groups (eGFR<15 ml/min/1.73 m2 vs eGFR≥15 ml/min/1.73 m2) regarding lipid arcs [14], but a study by Kato et al showed larger lipid arcs among CKD patients (eGFR<60 ml/min/1.73 m2) [15], similar to Sugiyama et al [71]. We found no statically significant difference in mean cap thickness and prevalence of TCFA between our 2 groups. In the study by Kato et al, fibrous cap thickness also did not differ between groups [15]. Moreover, an analysis in the PROSPECT study yielded similar results, showing that the prevalence of virtual histology-derived TCFA was not different [72]. In a study evaluating 296 coronary lesions among patients with stable angina treated with ad-hoc stent implantation, eGFR<15 ml/min/1.73 m2 was correlated with thinner fibrous cap and higher prevalence of plaque rupture [71]. We believe such results are not produced in our study because we excluded patients with severely deteriorated renal function. A study by Nakajima et al found that CKD is an independent clinical predictor for rapid plaque progression of calcifications [73]. Furthermore, after 6 months, OCT was conducted again and rapid progression of calcifications resulted in a significant reduction in the prevalence of TCFAs compared to 6 months prior [73]. Finally, a study enrolling 704 patients with ACS or CCS undergoing CAG assisted by IVUS and OCT found no independent association between eGFR<60 ml/min/1.73 m2 and IVUS- or OCT-derived plaque morphology [53].

In summary, most studies to date describing the relationship between impaired renal function and atherosclerotic plaque characteristics have focused on patients with advanced-stage CKD. They have demonstrated more frequent calcifications among CKD patients, which may require a different strategy (eg, high-pressure balloons, cutting balloons, intravascular lithotripsy, or rotablation) in case of PCI [20]. However, in our study such results were not presented; therefore, when performing a PCI in patients with mild-to-moderate renal dysfunction, one can expect a comparable number of calcifications to patients with normal renal function.

In the follow-up analysis, mild-to-moderate renal dysfunction was associated with increased all-cause mortality and a higher incidence of MACE during the 5-year observation. CKD is a well-established comorbidity that worsens prognosis [1,4–7,24,40,54,74].

The primary limitation of our study is a relatively small patient sample size and conducting it at only 1 site. Moreover, the study was restricted to only CCS and this data cannot be directly compared to patients with ACS. Several anatomical features (eg, bifurcations, ostial lesions, tortuous vessels) might have influenced the accuracy of OCT, FFR, and IVUS measurements. Moreover, when evaluating OCT or IVUS images, there might be heterogeneity between analysts, as the parameters are not computed by a program but the borders of different structures are marked by a human. Because the FFR result was decisive if PCI was performed, one can suspect that some patients may have received suboptimal treatment and high-risk plaques might have not been stented. Also, limited data were available in terms of MACE follow-up, and further studies on larger cohorts are warranted.

Conclusions

Mild-to-moderate renal dysfunction was not associated with significant differences in OCT- and IVUS-derived plaque morphology nor with functional indices characterizing intermediate-grade coronary stenoses among patients with CCS. Mild-to-moderate renal dysfunction was associated with higher risk of all-cause mortality and higher MACE prevalence in 5-year follow-up.