Coronary artery disease (CAD), as one of the leading causes of death in humans, is mainly caused by atherosclerosis (1). The formation of atherosclerotic plaque is known to be characterized by the accumulation of low-density lipoprotein cholesterol (LDL-C) and monocyte-derived macrophages in the arterial wall (2). Further plaque rupture can expose subendothelial collagen and cause an inflammatory response, which leads to platelet activation and coagulation cascade, resulting in thrombus formation (3). Thrombus can cause partial or complete occlusion of the coronary artery, leading to acute coronary syndrome (ACS) (4).

The number of circulating blood monocytes is closely associated with the formation and expansion of atherosclerosis in both human and animal models (5, 6). Monocytes can secrete enzymes that degrade extracellular matrix, contributing to the rupture of plaque underlying ACS (7). High-density lipoprotein (HDL), a cardioprotective factor, exhibits anti-atherosclerotic properties by neutralizing the pro-inflammatory and pro-oxidative effects of monocytes via inhibiting macrophage migration and LDL-C oxidation and mediating efflux of cholesterol from cells (8, 9). Of note, a recent Mendelian randomization study showed that low HDL cholesterol (HDL-C) was associated with high monocyte count (10). Therefore, the combination of monocyte and HDL-C may reflect the inflammatory and metabolic process of atherosclerosis better than the individual monocyte or HDL-C (9). The monocyte to HDL-C ratio (MHR) has been demonstrated to be associated with poor prognosis in many patient populations (11–19). The study of Cetin et al. including 2,661 patients with ACS showed that MHR was an independent predictor of CAD severity and future cardiovascular (CV) events (19). Nevertheless, the prognostic influence of MHR in patients with ACS undergoing percutaneous coronary intervention (PCI) has not been adequately studied and, importantly, few studies have evaluated the incremental value of adding MHR to the GRACE (Global Registry of Acute Coronary Events) risk score based prognostic models.

The objective of the present study was to investigate the relationship between MHR and CV outcomes in patients with ACS undergoing PCI, and then determine whether MHR could improve the GRACE risk score based prognostic models.

Of the 1,720 patients (mean age 60 ± 10 years), 401 (23.3%) were female and 1,319 (76.7%) were male. The baseline characteristics according to the MHR tertiles are shown in Table 1. Patients with higher MHR tertiles were younger, and had lower levels of blood pressure, total cholesterol, HDL-C, and left ventricular ejection fraction (LVEF), but had higher levels of BMI, monocyte count, neutrophil count, lymphocyte count, NLR, hs-CRP, triglycerides, FPG, glycosylated hemoglobin, cardiac troponin I (cTnI), SYNTAX score, and GRACE risk score. Patients with higher MHR tertiles were more likely to be male, and had higher rates of smoking, diabetes, dyslipidemia, previous MI, PAD, heart failure, MI, left main and/or multivessel lesions, and chronic total occlusion, but had lower rates of complete revascularization. In correlation analysis, MHR was significantly and positively correlated with hs-CRP (r = 0.412, P < 0.001), neutrophil to lymphocyte ratio (r = 0.203, P < 0.001), SYNTAX score (r = 0.115, P < 0.001), and GRACE risk score (r = 0.226, P < 0.001).

During the median follow-up of 31 months (IQR: 31–36 months), 353 patients developed at least one primary endpoint event, which was found in 80 (14.1%) patients from the T1 group, 121 (20.9%) from the T2 group, and 152 (26.5%) from the T3 group. The number and percentage of each component of the primary endpoint were as follows: 44 (2.5%) deaths, 49 (2.8%) MIs, 25 (1.4%) strokes, and 287 (16.3%) unplanned repeated revascularizations. Detailed clinical outcomes among patients with MHR tertiles are shown in Table 2. As shown in Supplementary Table 1, patients with the primary endpoint had higher levels of MHR, blood pressure, monocyte count, neutrophil count, NLR, hs-CRP, total cholesterol, triglycerides, LDL-C, FPG, glycosylated hemoglobin, and cTnI, but had lower LVEF and HDL-C. Those patients also had higher rate of diabetes, previous MI, prior PCI, CKD, PAD, and heart failure. In terms of the angiographic findings, patients with the primary endpoint were associated with higher rate of left-main and/or multivessel lesions, chronic total occlusion, lesions longer than 20 mm, bifurcation or trifurcation lesions and had higher SYNTAX score but lower rate of complete revascularization.

In ROC analysis, a MHR cut-off value of 9.9 had 57.5% sensitivity and 59.2% specificity for prediction of the primary endpoint (AUC = 0.594, 95% CI: 0.562–0.627, P < 0.001; Figure 1A). Moreover, a MHR cut-off value of 8.9 had 75.0% sensitivity and 46.9% specificity for prediction of the hard endpoint (AUC = 0.625, 95% CI: 0.571–0.680, P < 0.001; Figure 1B). Kaplan-Meier survival analysis showed that the cumulative incidence of the primary endpoint increased with higher MHR tertiles (log-rank test, P < 0.001; Figure 2). The difference in the cumulative incidence of the primary endpoint was mainly driven by an increase in overall death (log-rank test, P = 0.021), non-fatal MI (log-rank test, P = 0.018), and unplanned repeat revascularization (log-rank test, P < 0.001). However, the incidence of non-fatal stroke among the MHR tertiles were similar (log-rank test, P = 0.471; Figure 3).

Univariate and multivariate Cox proportional hazards regressions for the primary endpoint are shown in Table 3 and Supplementary Table 2, where the MHR were considered as categorical variable and continuous variable, respectively. In the univariate analyses, patients in the higher MHR tertiles were at higher risk of the primary endpoint [HR: 1.54 (95% CI: 1.16–2.04) for T2 and 2.03 (95% CI: 1.55–2.67) for T3]. After adjusting for other confounding factors, MHR across the tertiles could independently predict the occurrence of primary endpoint [adjusted HR: 1.45 [95% CI: 1.08–1.95) for T2 and 1.76 (95% CI: 1.30–2.40) for T3]. The incidence of the primary endpoint was monotonically increased across the MHR tertiles (P for trend < 0.001). When considering as a continuous variable, MHR was associated with an HR of 1.04 (95% CI: 1.02–1.06, P < 0.001) in the univariate analysis and with an HR of 1.03 (95% CI: 1.01–1.05, P = 0.004) in the multivariate analysis. In terms of the hard endpoint, the multivariate Cox proportional hazards regression analysis adjusted for multiple confounding factors revealed a hazard ratio for the hard endpoint of 2.34 (95% CI 1.27–4.33; P for trend = 0.023) when the highest and lowest MHR tertiles were compared. Subgroup analyses of MHR as a continuous variable for the primary endpoint were performed according to sex, age, hypertension, diabetes, dyslipidemia, type of ACS, and medications at discharge (Figure 4). One unit increase of MHR had a significant predictive role for different subgroups regardless of male or female, age ≥ or <60 years, hypertension or not, diabetes or not, dyslipidemia or not, STEMI or NSTE-ACS (unstable angina + NSTEMI), ACEI/ARBs use or not at discharge, β-blockers use or not at discharge (all P for interaction > 0.05).

The addition of MHR had an incremental effect on the predictive ability of the GRACE risk score for the primary endpoint (C-statistic: GRACE risk score + MHR vs GRACE risk score: 0.590 vs. 0.525, P < 0.001; cNRI: 0.136, P < 0.001; IDI: 0.006, P < 0.001; Table 4). Moreover, the addition of MHR could significantly increase the C-statistics of the GRACE risk score for the composite of death or MI and the composite of death, stroke, or MI (GRACE risk score + MHR vs. GRACE score: 0.648 vs. 0.605, P = 0.003, and 0.667 vs. 0.622, P = 0.023, respectively).

The present study investigated the relationship between MHR and adverse CV outcomes in patients with ACS undergoing PCI. The results showed that even after adjusting for as many confounding factors as possible, MHR remained significantly associated with the primary endpoint. Furthermore, the addition of MHR significantly improved the predictive ability of the GRACE risk score for adverse CV outcomes.

Atherosclerosis, as the main cause of CAD, is considered as an inflammatory disease. Inflammatory responses have been shown to be significantly enhanced in both obstructive and non-obstructive acute myocardial infarction patients accompanied with hyperglycemia (22). The inflammatory process is dominated by monocyte-derived macrophages in the early stage of plaque formation. Tissue-infiltrating macrophages eventually become the primary immune cells of plaque, where they take up cholesterol and store lipids in the form of small droplets, resulting in a unique morphology called foam cells (23). Monocyte recruitment governs the expansion of plaque formation, and the inhibition of monocyte infiltration and differentiation can attenuate early atherogenesis (24). Monocyte transmigration is induced by LDL but inhibited by HDL (25). Statins that used to lower LDL-C can reduce the number of monocyte-derived macrophages and the levels of C-reactive protein, eventually leading to plaque regression (26). Additionally, monocytes are associated with plaque rupture by secreting a variety of enzymes such as matrix metalloproteinases which are involved in the breakdown of the arterial extracellular matrix (27). HDL has been demonstrated to protect against plaque formation via promoting the efflux of cholesterol from cells and decreasing cholesterol levels (28). Moreover, HDL has anti-inflammation and anti-oxidation properties by reducing lipid peroxidation and deposition, and minimizing the accumulation of foam cells in the artery wall (29). The intrinsic relationship between monocytes and HDL in the development of atherosclerosis suggests that the combination of monocytes and HDL-C (presented as MHR) may better reflect the inflammatory process of atherosclerosis than either alone.

Indeed, MHR reflects both immune and metabolic status and is associated with pathological processes in many diseases (30–34). In patients with atherosclerotic CV disease mainly characterized by inflammation and lipid deposition, MHR has a significant prognostic role. It was reported that a higher MHR was associated with a worse prognosis in CAD patients who underwent PCI (14). MHR was shown to be an independent predictor of in-stent restenosis and adverse CV outcomes in patients with STEMI (35–37). Also, a higher MHR was demonstrated to be independently associated with a higher risk of intracoronary thrombus burden and no-reflow phenomenon (38, 39). Similar to these studies, we also found MHR was an independent predictor of adverse CV outcomes in patients with ACS undergoing PCI. Of note, there was a borderline differential effect of MHR across sex in our study. MHR seemed to have better predictive value of cardiovascular outcomes in female patients than in male patients. As we known, clinical outcomes from ACS are worse for women than for men (40). In our study, the correlation of MHR with hs-CRP was more significant in female patients (r = 0.424, P < 0.001) than in male patients (r = 0.418, P < 0.001). The hs-CRP is considered an ideal indicator of systemic inflammation and has been shown to be strongly associated with poor prognosis (41). This may be one reason why there was a borderline differential effect of MHR across sex. However, prospective studies with large sample sizes and sufficient statistical power are needed to confirm sex differences in MHR prediction of adverse cardiovascular events.

The GRACE risk score derived from an international registry involving more than 100,000 patients in 30 countries has been widely used to predict in-hospital and long-term outcomes in ACS patients. Many studies have confirmed the short- and long-term predictive value of the GRACE risk score (42, 43). However, inflammatory and lipid biomarkers were not taken into consideration in the GRACE risk score. In terms of inflammatory biomarkers, previous studies have shown that C-reactive protein has a weak correlation with the GRACE risk score and the addition of C-reactive protein can improve the predictive ability of the GRACE risk score (44). Also, the introduction of neutrophil counts or neutrophil to lymphocyte ratio can increase the predictive value of the GRACE risk score (45). Remarkably, monocyte counts were shown to be correlated with the GRACE risk score (46). Considering lipid biomarkers, the addition of LDL-C or lipoprotein(a) has been demonstrated to improve the predictive ability of the GRACE risk score (47, 48). Therefore, inflammatory and lipid biomarkers can add the predictive value on top of the GRACE risk score. In the present study, we found that adjustment of the GRACE risk score by MHR (as a marker for inflammation and lipid metabolism) improved prediction of adverse CV outcomes in patients with ACS undergoing PCI.

Several limitations must be taken into account when interpreting the results of our study. First, the present study was only a single-center observational study, so the effects of unmeasured and undetected confounding variables cannot be excluded. Second, the data of this study were limited to the Chinese population, so the ethnic difference cannot be eradicated. Third, both monocytes and HDL-C were measured at admission; however, values measured at discharge or changes during follow-up may be even more predictive.

MHR was independently and significantly associated with adverse CV outcomes and improved the predictive ability of the GRACE risk score in ACS patients who underwent PCI. Clinical trials are needed to determine whether medical management optimization based on the MHR reduces the risk of subsequent CV events.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by local Ethics Committee of Beijing Anzhen Hospital, Capital Medical University. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.

XM and KH analyzed the data and drafted the manuscript. XM, DS, and YZ designed the study and revised the manuscript. All authors contributed to the acquisition of data and final approval of the version to be published.

This work was supported by National Key Research and Development Program of China (2017YFC0908800), China Postdoctoral Science Foundation (2021M692253), Beijing Postdoctoral Research Foundation (2021-ZZ-023), and Beijing Municipal Administration of Hospitals Mission Plan (SML20180601).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.