Generally, low-density lipoprotein cholesterol (LDL-C) levels above a certain threshold are a well-known risk indicator for contributing to atherosclerotic cardiovascular disease (ASCVD). Lowering such levels is a paramount therapeutic goal of the guidelines drafted for managing hypercholesterolemia [1, 2]. However, in high triglyceride levels and metabolic diseases such as obesity, metabolic syndrome, gout, and diabetes mellitus, some individuals who maintain desirable LDL-C values still encounter the exacerbation of atherosclerosis, which may eventually lead to cardiovascular events [3, 4, 5]. Additionally, non-high-density lipoprotein cholesterol (non-HDL-C) can contribute to the atherogenic risk caused by various atherogenic risk components of remnant lipoprotein particles and is an acceptable surrogate marker for apolipoprotein B [3, 6, 7]. An increasing body of evidence, supported by earlier studies, has demonstrated that measuring non-HDL-C levels is superior to quantifying LDL-C in identifying statin-treated individuals with a higher residual risk of ASCVD who may require more intensive therapy; such measurements are already recommended by treatment guidelines [4, 8, 9, 10]. Many studies have discovered that a certain proportion of individuals with optimal LDL-C levels exhibit unexpectedly high non-HDL-C levels, coupled with high triglyceride levels, or metabolic syndrome and diabetes [5, 11]. Previous studies have postulated that the phenomenon of high non-HDL-C with low LDL-C is termed as the discordance and may reflect a greater residual ASCVD risk, regardless of whether an individual is undergoing lipid-lowering therapy [8, 10].

Patients with previous coronary artery bypass grafting (CABG) surgery experience an accelerated progression of atherosclerosis, which can further increase the incidence of recurrent cardiovascular (CV) events [12, 13, 14]. A previous study has discovered a better predictive value of non-HDL-C in terms of CV risk than LDL-C levels in this population. However, most of the participants in that study were not undergoing routine statin treatment [15]. Therefore, few studies have investigated whether a discordance of non-HDL-C levels and LDL-C levels correlated with elevated CV risk in statin-treated individuals with previous CABG compared with the risk in those with concordant levels of the two factors. Given these contemplations, the first objective of this study is to examine the association of serum concentrations of non-HDL-C with CV events among post-CABG individuals who received secondary percutaneous coronary intervention (PCI) treatment. The second objective was to determine whether a discordance between non-HDL-C and LDL-C levels affects prognosis.

The primary findings of this retrospective study shed some valuable insight on the predictive probability of non-HDL-C levels for evaluating the residual risk of long-term outcomes in ACS individuals with a history of CABG and statin therapy. The measurement of non-HDL-C levels better reflects the residual risk of CV events in this population than LDL-C levels. Interestingly, a discordance in lipid levels involving elevated non-HDL-C and low LDLC levels, but not one involving low non-HDL-C and high LDL-C levels, was significantly in association with an elevated likelihood of MACEs among ACS patients with previous CABG and statin treatment that underwent secondary coronary interventions.

A large body of previous studies has assessed the ability of non-HDL-C regarding the prediction of cardiac mortality and other events for the prevention of CV disease [3, 19, 20]. The increased concentration of small dense LDL particles, a reduction in HDL-C levels, the presence of hypertriglyceridemia, an increase in remnant lipoproteins, and postprandial hyperlipidemia were shown to be factors that partially accounted for the mechanisms underlying the residual CV risk observed in some individuals with optimal LDL-C levels [21]. These findings revealed that an increase in non-HDL-C levels indicated the residual risk of MACEs in statin-treated, ACS individuals and previous CABG. Comparable findings were reported by Fukushima et al. [15], which indicated that individuals with increased non-HDL-C levels following CABG struggled with significantly higher odds of CV death and other adverse clinical outcomes. However, that study failed to consider the risks associated with previous statin therapy. Similar findings from other studies have manifested that the possibility of serum non-HDL-C for the prediction of clinical prognosis was steadier and more enduring than other lipid parameters [8, 10]. On the other hand, Nakamura Y et al. [22] indicated that optimal medical treatments combined with statin treatment, antiplatelet agent, and beta-blockers reduced all-cause death and cardiac death in patients with a history of CABG who underwent PCI. However, in this study, we report a linear curve between the incidence of long-term cardiac events and non-HDL-C concentrations in the setting of moderate-intensity statin therapy for post-CABG patients that required PCI treatment. More specifically, non-HDL-C concentrations exceeding 2.6 mmol/L were likely to be powerful and accurate risk indications of CV prognosis, similar to the results reported by Brunner et al. [19]. According to our results from the subgroup analysis, the threshold concentration of non-HDL-C as a risk modifier in statin-treated individuals with optimal LDL-C values would be closer to 2.07 mmol/L. For the participants with undesirable LDL-C levels, only a relatively moderate association was observed between non-HDL-C levels and the risks of CV mortality, recurrence of MI, and the need for revascularization. Thus, achieving target LDL-C concentrations in those patients remains the primary goal for reducing risk in the initial management of patients with previous CABG. These findings show that evaluating non-HDL-C could represent a secondary means of appraising residual risk regarding CV events and identifying extremely high-risk patients with a history of CABG who may require intensification of lipid-lowering therapies.

Patients with a history of CABG typically exhibit a complicated and diffuse progression of atherosclerosis and other complications (e.g., ischemia stroke, diffuse peripheral arterial atherosclerosis, serious renal insufficiency, and rheumatology disease) and struggle with more fatal and non-fatal events [23, 24, 25]. These findings indicated in this population that discordance between low LDL-C and high non-HDL-C levels was observed in up to 5.1% of cases, similar to the proportions reported by the previous studies [26, 27]. In a recent Johannesen et al. [10] study, a discordance involving low LDL-C/ high non-HDL-C or apo(B) was associated with a 91% higher risk of MI and a 23% higher risk of mortality compared with the risk in those with concordantly low levels of both factors; a similar trend was not observed in those exhibiting a discordance involving high LDL-C and low non-HDL-C levels. Several explanations have been provided for this phenomenon, which is prevalently consistent with the previous studies [8, 10]. Firstly, non-HDL cholesterol fraction encompasses circulating LDL-C, very low-density lipoprotein (VLDL) cholesterol, and other ingredients from byproducts of triglyceride-related lipoprotein metabolism. Increased levels of atherogenic cholesterol are superior indicators of the mass of lipoprotein particles in evaluating associations that affect ASCVD risk [3]. This discordance may reflect higher concentrations of remnant cholesterol, which could lead to an atherosclerotic cardiovascular disease beyond apolipoprotein B and the significant amounts of atherogenic LDL particulates that interact with the coronary artery. It is not enough to measure the concentrations of these circulating lipoprotein components [5, 28]. In addition, the discordance between two cholesterol parameters was mediated by metabolic syndrome and unrelated to the conventional risk factors regardless of the BMI and the degree of glycemic control [5, 10], as the findings of Model 3 were indicative of the absence of multicollinearity. In patients exhibiting a discordance involving low LDL-C and high apolipoprotein B or non-HDL-C levels, an increased risk of arterial stiffness was observed, as measured by brachial-ankle pulse wave velocity; these factors are regarded as markers of subclinical atherosclerosis [27, 29]. In general, elevated arterial stiffness and multifocal atherosclerosis in patients after CABG tracked more with the atherosclerosis progression and a poor prognosis [30, 31, 32]. Although the biological mechanisms underlying the causality of these relationships remain inexplicable, those exhibiting a discordance between two lipid parameters presented elevated levels of remnant TGs or cholesterol related to insulin resistance. In turn, this contributed to increased arterial stiffness and the development of diffuse and multifocal atherosclerosis because of the disorder of intimal cells caused by varying degrees of oxidative responses and impaired endothelial function [11, 33]. In this study, the subgroup of patients with a CABG history with elevated non-HDL-C and optimal LDL-C levels exhibited a risk of CV death or re-infarction that was as much as four times higher than that observed in the group with concordant lipid levels confirming that further efforts are needed in such patients to ensure enhanced low-lipid management.

In addition to the inevitable selection bias inherent to retrospective studies, several limitations existed in this study. First, the small example size and relatively low incidence of CV death/re-infarction in the enrolled patients may have contributed to a high selection bias that could complicate the interpretation of the results. A comparable prevalence of CV deaths is reported in another real-world Chinese research [34]. Thus, it is possible that any inconsistencies could be related to the ethnic makeup of the group and could provide some insight into the predictive value of lipid discordance on prognosis in Chinese patients with a history of CABG. Second, apolipoprotein B measurements were not performed in this study because of the limitations of our laboratory; therefore, the analysis was limited to comparing groups based on LDL-C and non-HDL-C levels. Based on the high correlation between these two parameters, the Adult Treatment Panel III (ATP III) guidelines recommend using non-HDL-C levels as a reasonable substitute for apolipoprotein B concentrations [35]. Third, approximately half of the participants in this study did not reach the optimal LDL-C levels recommended by the guidelines, which could have influenced the credibility of the results. Therefore, a subgroup analysis was conducted that was stratified based on the median LDL-C concentration to classify the impact of non-HDL-C levels on prognosis. Large-scale research is necessary to explore the effect of lipid discordance on saphenous grafting and clinical prognosis in post-CABG patients with statin treatment who underwent PCI treatment.

In post-CABG and statin treated ACS individuals, who received secondary PCI, there is a linear association of non-HDL-C with the significant risk of MACE. Moreover, a linear relationship between non-HDL-C values exceeding 2.07 mmol/L and risks of CV death/recurrent MI was presented in those patients with desirable LDL-C levels ($\le$1.8 mmol/L). The discordance of high non-HDL-C/low LDL-C could provide utility in identifying the residual risk of MACEs in this population.

MACE, major adverse cardiovascular events; CAD, coronary artery disease; PCI, percutaneous coronary intervention; GRACE, Global Registry of Acute Coronary Events; MI, myocardial infarction; CABG, coronary artery bypass grafting; TC, total cholesterol; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; TG, triglyceride; Lp(a), apolipoprotein A; CCR, creatinine clearance rate; ESR, erythrocyte sedimentation rate; Hs-CRP, high-sensitivity C-reactive protein; cTnI, cardiac troponin I; CKMB, creatine kinase–myocardial band; BMI, body mass index; LVEF, left ventricular ejection fraction; ROC, receiver operating characteristic; AUC, area under the ROC.

The datasets generated and analyzed are not publicly available due to the policies of Beijing Chaoyang Hospital regarding individual confidentiality; however, they are available from the corresponding author upon reasonable request.

CL, KZH and YXY designed the present study, conducted data analysis and drafted the manuscript. KBL, MLC and LFW gave critical revision opinions on the manuscript drafts, provided material and technical supported during data analysis. XRX and YFG aided interpretation of data, commented on this study design and provided critical review. All authors have read and approved the manuscript. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was approved by the institutional review board of Beijing Chaoyang Hospital (2017-S-187) and performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. Written informed consent forms were waived due to the nature of the retrospective study.

Not applicable.

This research received no external funding.

The authors declare no conflict of interest.