Abstract
Abstract
Objectives
To develop a CERT SCORE utilising both odd-chain and even-chain ceramide species, and to evaluate its association with short-term and lifetime cardiovascular risk in patients with coronary atherosclerotic heart disease (CAD).
Design
Prospective cohort study.
Setting
A patients-based cohort in the Beijing Anzhen Hospital, Capital Medical University.
Participants
CAD patients were defined as having at least one coronary artery stenosis of ≥50% as assessed by coronary angiography or CT angiography.
Main outcomes and measures
Major adverse cardiac events (MACE) including all-cause death, myocardial infarction, heart failure, cerebral infarction and readmission.
Results
We quantified 13 ceramide species and calculated the ratios of Cer(d18:1/14:0) and Cer(d18:1/24:0). Based on these measurements, Cer(d18:1/19:0), Cer(d18:1/19:0)/Cer(d18:1/14:0), Cer(d18:1/19:0)/Cer(d18:1/24:0) and Cer(d18:1/21:0)/Cer(d18:1/24:0) were selected to construct the CERT SCORE. Using this score, patients were classified into two distinct risk categories for MACE: low-risk (score 0–6) and high-risk (score 7–12). The high-risk group exhibited a significantly higher short-term risk of MACE (HR 2.10; 95% CI 1.50 to 2.94) compared with the low-risk group. The cumulative MACE risk in the low- and high-risk groups during the 1000-day follow-up was 25.45% and 44%, respectively. Subgroup analyses revealed that the presence of multivessel coronary artery lesions did not significantly modify the association between the CERT SCORE and short-term MACE risk (p value for interaction=0.967). Furthermore, in the age groups of 41–50 years, 51–60 years and 61–70 years, lifetime risk was significantly elevated in the high-risk group compared with the low-risk group.
Conclusion
We have developed a ceramide-based risk stratification tool (CERT SCORE) that demonstrates robust predictive value for identifying high-risk MACE patients among CAD patients. This tool offers considerable clinical utility for guiding patient management and informing therapeutic decisions.
Keywords: Ischaemic heart disease, Cardiac Epidemiology, Coronary heart disease, Angina Pectoris, Cardiovascular Disease
STRENGTHS AND LIMITATIONS OF THIS STUDY.
This is the first study that incorporates odd-chain ceramides into the prognostic model for risk prediction in patients with coronary atherosclerotic heart disease (CAD).
This is the first investigation to use ceramide profiling for lifetime risk assessment in CAD patients.
The inability to distinguish between cardiovascular and non-cardiovascular causes of death during follow-up may have introduced outcome misclassification bias.
The lack of adjustment for the relationship between medication use and ceramide levels may represent another source of bias in the study results.
Introduction
Coronary atherosclerotic heart disease (CAD) is a prevalent cardiovascular disease (CVD) worldwide, characterised by the formation and progression of atherosclerotic plaques in the coronary arteries, leading to varying degrees of luminal narrowing (from mild to severe) and even complete occlusion, thereby inducing myocardial ischaemia.1 2 Although significant progress has been made in diagnosis and treatment over the past decade, CAD continues to pose a major public health challenge, contributing to substantial economic burdens and persistently high mortality rates.3 Therefore, there is an urgent need for improved risk assessment tools to better identify high-risk CAD patients, which is essential for enhancing clinical care and optimising long-term disease management.4
Plasma ceramides have emerged as promising biomarkers for cardiovascular risk, offering predictive value beyond traditional plasma lipid biomarkers.5 6 In addition to their structural role in cell membranes,7 ceramides participate in key pathological processes such as inflammation, apoptosis and dysregulation of lipid metabolism.8,11 Studies have shown that plasma ceramide levels are significantly elevated in patients with CAD compared with healthy individuals.12 Beyond concentration, ceramide chain length is associated with the severity of atherosclerosis.13
Currently, a number of predictive models based on plasma ceramides, such as CERT1, CERT2 and CERT2-TnT, are being developed to assess the prognosis of patients with CAD.6 14 15 These models enable effective risk stratification at the individual level, thereby supporting clinical decision-making and optimising the allocation of medical resources. However, they only contain even-chained ceramides (Cer(d18:1/16:0), Cer(d18:1/18:0), Cer(d18:1/24:1) and Cer(d18:1/24:0)). Recently, there has been growing interest in odd-chain ceramides, which were associated with disorders of lipid metabolism16 and poorer outcomes in acute coronary syndrome (ACS)17 or acute myocardial infarction (MI).18 Thus, incorporating both odd-chain and even-chain ceramides into predictive models may further improve prognostic accuracy in patients with CAD. Additionally, existing models mainly target short-term risk and do not incorporate lifelong cardiovascular risk, which could restrict their usefulness in guiding long-term preventive strategies.
Therefore, this study aims to develop a novel ceramide score (CERT SCORE) by integrating both odd-chain and even-chain ceramides to predict short-term prognosis and evaluate lifetime cardiovascular risk in Chinese patients with CAD. This approach is expected to enhance patient management and improve cost-effectiveness in clinical practice.
Methods
Study population
CAD patients who underwent coronary angiography or computed tomography (CT) at Beijing Anzhen Hospital, Capital Medical University, between February 2016 and March 2018 were enrolled in this study. Patients with at least one coronary artery stenosis of ≥50% defined by coronary angiography or coronary CT were included in the study. The exclusion criteria were as follows: (1) unavailable blood sample; (2) a history of percutaneous coronary intervention or coronary artery bypass grafting; (3) patients with advanced heart failure, malignant tumours, severe infections, trauma, liver and kidney dysfunction or other major diseases and complications and (4) incomplete clinical information.
This study was approved by the Ethics Committee of Beijing Anzhen Hospital, Capital Medical University, and written informed consent was obtained from all participants. The research adhered to the tenets of the Declaration of Helsinki.
Ceramide standard
The ceramide external standards Cer(d18:1/12:0), Cer(d18:1/14:0), Cer(d18:1/16:0), Cer(d18:1/17:0), Cer(d18:1/18:0), Cer(d18:1/19:0), Cer(d18:1/20:0), Cer(d18:1/22:0), Cer(d18:1/24:1) and Cer(d18:1/24:0), along with the internal standards (IS): Cer(d18:1/16:0)-d7, Cer(d18:1/18:0)-d7, Cer(d18:1/24:1)-d7 and Cer(d18:1/24:0)-d7 were prepared as primary stock solutions of 100 mM. Subsequently diluted to secondary stock solutions of 10 mM and 1 mM. All were stored at −80°C. Cer(d18:1/19:0) was purchased from Cayman Chemical, and other standards were purchased from Avanti Polar Lipids. The concentrations for the standard curves were optimised based on pre-experimental results, with the IS concentration set at 0.1 µM. Cer(d18:1/15:0), Cer(d18:1/21:0) and Cer(d18:1/23:0) were quantified using the standard curves established of Cer(d18:1/12:0), Cer(d18:1/14:0) and Cer(d18:1/22:0), respectively.
Ceramide extraction
The plasma was thawed at 4°C for standby use. The IS stock solution was prepared in methanol, containing Cer(d18:1/16:0)-d7, Cer(d18:1/18:0)-d7, Cer(d18:1/24:1)-d7 and Cer(d18:1/24:0)-d7. To minimise exogenous lipid contamination, all lipid extraction steps were carried out using glass tubes (Fisher Scientific). 25 µL plasma was placed into a glass tube, followed by the addition of 300 µL of the methanol-based IS solution. The mixture was vortexed thoroughly and then centrifuged at 3500 rpm/min for 10 min. The resulting supernatant was evaporated to dryness under a gentle nitrogen stream at 37 °C. The residue was reconstituted in 50 µL of a 1:1 (v/v) methanol: acetonitrile solution, vortexed to ensure complete re-dissolution and subsequently transferred to a chromatographic vial. The details of quantification of ceramides were displayed in the online supplemental methods.
Outcomes
The outcome of this study was the occurrence of major adverse cardiac events (MACE), defined as a composite of all-cause death, MI, heart failure, cerebral infarction and readmission. All participants were followed up through telephone interviews, medical record reviews and civil registration to ascertain the occurrence of MACE.
Statistical analyses
The quantitative clinical characteristics of patients with and without MACE were calculated. Categorical variables were presented as numbers (percentages) and were compared using the χ2 test. Continuous variables were expressed as means [±SD] in normal distributions or as medians (IQR) and compared using unpaired Student’s t-test or Mann-Whitney U test, as appropriate.
Ceramide concentrations were log-transformed and standardised to a mean of zero and SD of one. Cox proportional hazard regression was used to calculate HR and 95% CIs to estimate the association between ceramides and MACE. Two multivariable models were conducted: model 1 adjusted for age, sex, body mass index, smoking, systolic blood pressure, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol (LDL-C), high-sensitivity C reactive protein (hs-CRP) and diabetes; model 2 further adjusted for the presence of coronary lesions, including left anterior descending, left circumflex, right coronary artery and left main. To account for multiple comparisons, a false discovery rate (FDR) <5% was applied for statistical significance.
To identify ceramides significantly and independently associated with MACE, two machine learning-based feature selection methods were applied: least absolute shrinkage and selection operator (LASSO) Cox regression and random survival forest (RSF). Ceramides meeting all of the following criteria were included to establish CERT SCOREs: (1) FDR<5% in model 2 of the Cox regression; (2) selected by LASSO Cox regression; (3) ranked among the top 10 most important features in the RSF. Finally, four ceramide-related variables were selected based on these criteria: Cer(d18:1/19:0), Cer(d18:1/19:0)/Cer(d18:1/14:0), Cer(d18:1/19:0)/Cer(d18:1/24:0) and Cer(d18:1/21:0)/Cer(d18:1/24:0). Each variable was assigned points from 0 to 3 according to its quartile distribution (online supplemental table 1). The CERT SCORE was then calculated as the sum of points across all four variables, resulting in a total score ranging from 0 to 12. Based on this score, Patients were categorised into two MACE risk categories: low-risk (score 0–6) and high-risk (score 7–12).
The Kaplan-Meier method was used to calculate the cumulative risk of MACE among CAD patients with CERT SCORE of 0–6 and 7–12 over 3 years. After adjusting for covariates in model 2, Cox proportional hazards regression was performed to evaluate the independent role of the CERT SCORE in the development of MACE in CAD patients.
A modified Kaplan-Meier method was employed to estimate the lifetime risk of MACE in CAD patients stratified by CERT SCORE (0–6 vs 7–12).19 The lifetime risk of MACE up to 80 years of age was calculated for each index age ranging from 45 years to 60 years, and the lifetime risk of MACE at the index ages of 45 years, 55 years and 65 years represented the age groups 41–50 years, 51–60 years and 61–70 years, respectively. The age of 80 years was defined as the end of the lifetime span due to a mean life expectancy of 77.8 years in 2020 in China,20 which represents most of the Chinese lifetime.
A joint analysis was performed to evaluate the interaction between the CERT SCORE and multivessel coronary artery lesions (MCAL). Patients were stratified into four categories according to their CERT SCORE and MCAL status: low-risk without MCAL, low-risk with MCAL, high-risk without MCAL and high-risk with MCAL. The 3-year and lifetime MACE risk of the four patient categories were calculated to evaluate the modifying effect of MCAL on the association of the CERT SCORE with short-term and lifetime MACE risk among CAD patients.
All analyses were conducted using R V.4.3.3 software and associated packages. A two-sided p value of <0.05 on the two-sided test was considered statistically significant unless otherwise specified.
Patient and public involvement
Patients or the public were not involved in the design, conduct, reporting or dissemination plans of our research.
Results
Baseline characteristics
A total of 632 patients were included in this study with a mean (±SD) age of 61.9±9.7 years, and 27.8% were female (table 1). During a median follow-up of 565 days, 171 patients (27.1%) experienced MACE, and they were more likely to have lower levels of diastolic blood pressure (p=0.008) and higher hs-CRP levels (p=0.041). When compared with patients without MACE, patients with MACE had a higher prevalence of MCAL (70.8% vs 62.3%, p=0.058) at baseline. Other clinical characteristics were not observed to be significantly different between patients with and without MACE (all p>0.05).
Table 1. Baseline characteristics of the study patients.
| Overall | Non-MACE | MACE | P value | |
|---|---|---|---|---|
| N | 632 | 461 | 171 | |
| Age, years | 61.88 (9.69) | 61.46 (9.77) | 63.01 (9.41) | 0.074 |
| Female, n (%) | 176 (27.8) | 125 (27.1) | 51 (29.8) | 0.565 |
| Hypertension, n (%) | 403 (63.8) | 297 (64.4) | 106 (62.0) | 0.636 |
| SBP, mm Hg | 138.61 (24.82) | 139.17 (25.50) | 137.11 (22.89) | 0.356 |
| DBP, mm Hg | 81.53 (15.12) | 82.50 (15.46) | 78.91 (13.89) | 0.008 |
| Hyperlipidaemia, n (%) | 296 (46.8) | 210 (45.6) | 86 (50.3) | 0.332 |
| Smoking, n (%) | 307 (48.6) | 222 (48.2) | 85 (49.7) | 0.797 |
| Diabetes, n (%) | 214 (33.9) | 149 (32.3) | 65 (38.0) | 0.212 |
| Height, m | 1.67 (0.08) | 1.67 (0.08) | 1.67 (0.07) | 0.87 |
| Weight, kg | 72.44 (11.54) | 72.62 (11.71) | 71.94 (11.09) | 0.507 |
| BMI, kg/m2 | 25.96 (3.16) | 26.04 (3.28) | 25.72 (2.80) | 0.258 |
| TG, mmol/L | 1.38(0.99, 1.99) | 1.37(0.97, 2.02) | 1.38(1.00, 1.90) | 0.697 |
| TC, mmol/L | 4.14 (0.98) | 4.12 (0.96) | 4.17 (1.02) | 0.618 |
| HDL-C, mmol/L | 1.11 (0.27) | 1.12 (0.28) | 1.10 (0.23) | 0.28 |
| LDL-C, mmol/L | 2.36 (0.78) | 2.33 (0.73) | 2.45 (0.88) | 0.08 |
| hs-CRP, mg/L | 1.46(0.58, 3.32) | 1.39(0.53, 3.20) | 1.78(0.72, 4.08) | 0.041 |
| FBG, mmol/L | 6.67 (2.41) | 6.59 (2.30) | 6.91 (2.66) | 0.139 |
| LM lesion, n (%) | 47 (7.4) | 36 (7.8) | 11 (6.4) | 0.678 |
| LAD lesion, n (%) | 519 (82.1) | 375 (81.3) | 144 (84.2) | 0.473 |
| LCX lesion, n (%) | 348 (55.1) | 243 (52.7) | 105 (61.4) | 0.063 |
| RCA lesion, n (%) | 367 (58.1) | 258 (56.0) | 109 (63.7) | 0.095 |
| MCAL, n (%) | 408 (64.6) | 287 (62.3) | 121 (70.8) | 0.058 |
Data are expressed as number (percent) for categorical variables and mean±SD for continuous variables in normal distributions.
BMI, body mass index; DBP, diastolic blood pressure; FBG, fasting blood-glucose; HDL-C, high-density lipoprotein cholesterol; hs-CRP, high-sensitivity C reactive protein; LAD, left anterior descending; LCX, left circumflex; LDL-C, low-density lipoprotein cholesterol; LM, left main; MACE, major adverse cardiac events; MCAL, multivessel coronary artery lesions; RCA, right coronary artery; SBP, systolic blood pressure; TC, total cholesterol; TG, triglycerides.
Association of ceramides with MACE
Cox proportional hazards regression models were used to estimate the relationship between MACE and 37 traits (figure 1 and online supplemental table 2), including 13 ceramides and their ratios to Cer(d18:1/14:0) and Cer(d18:1/24:0). After adjusting for the traditional risk factors of CVD, 19 traits showed a significant association with MACE risk (p<0.05). Among these 19 traits, Cer(d18:1/14:0) and Cer(d18:1/14:0)/ Cer(d18:1/24:0) exhibited a positive association with MACE, while the remaining 17 traits showed a negative association (model 1 in figure 1). After further adjustment for presence of coronary lesions (model 2 in figure 1), these relationships changed slightly, and 15 traits had an FDR<5%. The relationships of all 37 traits with MACE were evaluated using LASSO (online supplemental figure 1A) and RSF (online supplemental figure 1B). Using all three methods (COX regression, LASSO and RSF), four traits were consistently identified as being significantly associated with MACE risk (online supplemental figure 2): Cer(d18:1/19:0), Cer(d18:1/19:0)/Cer(d18:1/14:0), Cer(d18:1/19:0)/Cer(d18:1/24:0) and Cer(d18:1/21:0)/ Cer(d18:1/24:0). The CERT SCORE was constructed based on these four traits (online supplemental table 1).
Figure 1. Cox proportional hazards regression analysis examined the association between ceramide species and MACE in CAD patients. Model 1: adjusted for age, sex, BMI, smoke, SBP, TG, HDL-C, LDL-C, hs-CRP and diabetes. Model 2: additionally adjusted for LAD, LCX, RCA and LM lesion. BMI, body mass index; CAD, coronary artery disease; HDL-C, high-density lipoprotein cholesterol; hs-CRP, high-sensitivity C reactive protein; LAD, left anterior descending; LCX, left circumflex; LDL-C, low-density lipoprotein cholesterol; LM, left main; MACE, major adverse cardiac events; RCA, right coronary artery; SBP, systolic blood pressure; TG, triglycerides.
CERT SCORE and short-term risk of MACE
All patients were divided into low-risk (0–6) or high-risk (7–12) groups according to the CERT SCORE. The cumulative MACE risk in the low- and high-risk groups during the 1000-day follow-up was 25.45% and 44%, respectively (figure 2A). Compared with those in the low-risk group, the patients in the high-risk group had a 110% increased risk of MACE (HR 2.10, 95% CI 1.50 to 2.94). Subsequently, in the joint analysis of CERT SCORE and MCAL, we found that, compared with the low-risk group without MCAL, those in the high-risk group with and without MCAL had a 128% (HR 2.28, 95% CI 1.28 to 4.07) and 110% (HR 2.10, 95% CI 1.17 to 3.74) increased risk of MACE, respectively (figure 2B). However, patients with MCAL in the low-risk group did not exhibit a higher risk of MACE (HR 1.09, 95% CI 0.60 to 1.98). The subgroup analysis of MCAL indicated that high levels of CERT SCORE were associated with a higher risk of MACE both in patients with (HR 1.93, 95% CI 1.30 to 2.87) and without MCAL (HR 3.00, 95% CI 1.52 to 5.92). MCAL did not have a modifying effect on the relationship between CERT SCORE and MACE (p value for interaction=0.967).
Figure 2. Short-term risk in different groups. (A) Not considering MCAL, (B) Considering MCAL. CERT SCORE, ceramide score; MCAL, multivessel coronary artery lesions.
CERT SCORE and lifetime risk of MACE
Figure 3 presents the lifetime risk of MACE up to the age of 80 years among patients in the low-risk and high-risk groups at age 41–50, 51–60 and 61–70 years, as well as those with and without MCAL. The lifetime risk of MACE up to age of 80 years for patients was 94.53% (95% CI 90.23% to 98.83%) in the low-risk group and 99.66% (95% CI 99.24% to 100%) in the high-risk group among patients aged 41–50 years (online supplemental table 3). Additionally, compared with the low-risk group, the patients in the high-risk group had a higher lifetime risk both with (99.83% vs 95.39%) and without (99.17% vs 94.78%) MCAL. Similar results were consistently observed in the 51–60 years and 61–70 years age groups.
Figure 3. Lifetime risk of MACE up to age 80 years for patients at age 45 years (A), 55 years (B) and 65 years (C) in low-risk and high-risk group and whether with or without MCAL (D~F). CERT SCORE, ceramide score; MACE, major adverse cardiac events; MCAL, multivessel coronary artery lesions.
Discussion
In this prospective cohort study, we used an ultra-performance liquid chromatography-mass spectrometry platform to quantify 13 plasma ceramides in CAD patients. We constructed a CERT SCORE using Cer(d18:1/19:0), Cer(d18:1/19:0)/Cer(d18:1/14:0), Cer(d18:1/19:0)/Cer(d18:1/24:0) and Cer(d18:1/21:0)/ Cer(d18:1/24:0). This score effectively stratified short-term MACE risk into high-risk and low-risk categories, which is more suitable for the Chinese population. To our knowledge, this is the first study to estimate the lifetime risk of MACE associated with ceramides, and we showed that lifetime risk stratification using the CERT SCORE remains robust and is not significantly influenced by the presence of MCAL. These findings support that CERT SCORE could be served as a risk assessment tool for MACE of CVD patients.
Ceramides are critically involved in multiple stages of atherosclerotic lesion development, including monocyte recruitment, adhesion to the vascular endothelium and subsequent differentiation into macrophages.21 They function as bioactive mediators in inflammatory signalling pathways and contribute to endothelial dysfunction by enhancing reactive oxygen species production and reducing nitric oxide bioavailability.22 23 Furthermore, ceramides play a central role in regulating apoptosis and smooth muscle cell (SMC) phenotype. Studies have shown that SMC number in atherosclerotic plaques is negatively correlated with ceramide content, while ceramide content is positively correlated with caspase-3 expression.24 The mechanisms of macrophage apoptosis triggered by ox-LDL, aSMase activity and ceramide accumulation have not been fully elucidated.21 Moreover, the effects of ceramides with different chain lengths in atherosclerosis are inconsistent and can be attributed to different subtypes of ceramide synthases. Future studies should focus on elucidating the chain-specific mechanisms of ceramides in atherosclerosis and developing therapeutic strategies that target ceramide metabolism to prevent or treat the disease.
Ceramides participate in many pathological processes in CVD and have emerged as promising biomarkers and tools for risk stratification. Ceramide concentrations and associated risk scores are elevated in individuals with acute MI,18 CAD,25 26 ACS27 and recurrent MACE.28 Rejio et al confirmed that ceramides are predictive markers of death in CVD patients independently of LDL-C. They established the CERT1 score based on Cer(d18:1/16:0), Cer(d18:1/18:0), Cer(d18:1/24:1) and their ratios to Cer(d18:1/24:0) and they showed that Cer(d18:1/24:0) is a protective factor.6 However, in our cohort, Cer(d18:1/24:0) did not show a protective effect; instead, Cer(d18:1/14:0) exhibited protective properties. Furthermore, in our cohort, Cer(d18:1/16:0) and Cer(d18:1/24:1) levels were not associated with MACE, which may be related to ethnicity. The CERT2 score, developed in the Western Norway Coronary Angiography Cohort, introduced phosphatidylcholine based on CERT1 and demonstrated a higher HR than CERT1 in the LIPID and KAROLA validation cohorts.15 Furthermore, in the STABILITY cohort, the CERT2 risk score was confirmed as a reliable indicator of risk in patients with stable CAD and could be used to detect residual lipid and inflammatory risk in these patients. It has demonstrated consistent performance across research populations from different geographical locations worldwide.29
CERT1 and CERT2 scores are independent of the traditional plasma lipid biomarker LDL-C, indicating that ceramides and LDL-C may represent two distinct biological pathways in CVD progression. Future investigations should focus on elucidating the specific mechanisms and interactions between these two biomarkers in CVD development. A deeper understanding of their interplay is essential for advancing our knowledge of CVD pathogenesis and facilitating the design of more effective prevention and treatment strategies.
Emerging evidence suggests that other sphingolipids beyond ceramides may play critical roles in the progression of CVD. Annelise et al analysed 32 sphingolipids and developed a comprehensive CAD risk score incorporating eight of these lipids.12 They suggested that more abundant ceramide species, even those already identified as drivers of tissue and metabolic dysfunction, may not necessarily represent the most sensitive biomarkers for CAD. Instead, they highlighted that less abundant lipid metabolites, which serve as markers of increased ceramide biosynthesis flux, may offer a more accurate and comprehensive assessment of disease status.12 Further investigation is warranted to explore how broader sphingolipid profiles, beyond plasma ceramides alone, may influence cardiovascular outcomes and therapeutic opportunities.
It can be observed that the short-term risk of MACE was significantly higher in the high-risk group than in the low-risk group. It is consistent with the subgroup analysis, confirming that the CERT SCORE is an effective short-term risk stratification tool for CAD patients and is not affected by the presence of MCAL. CAD is a chronic condition that is often closely associated with comorbidities such as diabetes and hyperlipidaemia, long-term management and risk assessment are essential.30 Therefore, we further evaluated the lifetime risk of CAD using the CERT SCORE. Consistent with the short-term outcomes, the lifetime risk was also elevated in the high-risk group compared with the low-risk group. However, CERT SCORE cannot effectively stratify the lifetime risk into high- and low-risk groups of CAD patients at ages 41–50 years and 51–60 years without MCAL and at ages 61–70 years with MCAL. This may be related to the small number of patients with MCAL in this cohort, so a larger sample size is needed to clarify the interplay between the CERT SCORE, MCAL and long-term clinical outcomes.
Our study has several limitations. First, the cause of death (cardiovascular vs non-cardiovascular) was not recorded during follow-up. Second, we did not adjust for medications such as metformin and statins, which are known to affect plasma ceramide concentrations and may have confounded the results. Third, the model lacked external validation, as it was developed and evaluated solely within a single cohort. Additionally, the relatively small sample size may limit the stability and generalisability of our findings. Specifically, the sample size was insufficient to ensure adequate statistical power for subgroup analyses or to robustly validate the prognostic performance of the CERT SCORE across diverse clinical subsets. This increases the risk of overfitting and may affect the reproducibility of the results in broader populations.
Conclusion
In this study, we developed the first prognostic model incorporating odd-chain ceramides for risk stratification in patients with CAD. Our results suggest that CAD patients with a higher CERT SCORE are associated with an increased short-term and lifetime risk of MACE, indicating its potential utility as a prognostic tool. Future studies with larger, multicentre cohorts and external validation are necessary to confirm the clinical applicability of the CERT SCORE. Further investigation is also warranted to elucidate the roles of specific ceramide species in CAD progression, which may contribute to a more comprehensive understanding of cardiovascular pathogenesis and inform targeted therapeutic strategies.
Supplementary material
Footnotes
Funding: This study was supported by the National Natural Science Foundation of China (81970392, 82370440).
Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-094100).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and the study was approved by the Ethics Committee of Beijing An Zhen Hospital, Capital Medical University, with the approval number of 2024170x. Participants gave informed consent to participate in the study before taking part.
Patient and public involvement: Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Data availability statement
Data are available upon reasonable request.
References
- 1.Hermida RC, Ayala DE, Mojón A, et al. Influence of time of day of blood pressure-lowering treatment on cardiovascular risk in hypertensive patients with type 2 diabetes. Diabetes Care. 2011;34:1270–6. doi: 10.2337/dc11-0297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Šerpytis R, Puodžiukaitė L, Petrauskas S, et al. Outcomes of a percutaneous coronary intervention versus coronary artery bypass grafting in octogenarians. Acta Med Litu. 2018;25:132–9. doi: 10.6001/actamedica.v25i3.3860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Quirós-Fernández R, López-Plaza B, Bermejo LM, et al. Oral Supplement Containing Hydroxytyrosol and Punicalagin Improves Dyslipidemia in an Adult Population without Co-Adjuvant Treatment: A Randomized, Double-Blind, Controlled and Crossover Trial. Nutrients. 2022;14:1879. doi: 10.3390/nu14091879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hippisley-Cox J, Coupland C, Brindle P. Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. BMJ. 2017;357:j2099. doi: 10.1136/bmj.j2099. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Leandro AC, Michael LF, Almeida M, et al. Influence of the Human Lipidome on Epicardial Fat Volume in Mexican American Individuals. Front Cardiovasc Med. 2022;9:889985. doi: 10.3389/fcvm.2022.889985. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Laaksonen R, Ekroos K, Sysi-Aho M, et al. Plasma ceramides predict cardiovascular death in patients with stable coronary artery disease and acute coronary syndromes beyond LDL-cholesterol. Eur Heart J. 2016;37:1967–76. doi: 10.1093/eurheartj/ehw148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hannun YA, Obeid LM. Sphingolipids and their metabolism in physiology and disease. Nat Rev Mol Cell Biol . 2018;19:175–91. doi: 10.1038/nrm.2017.107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kovilakath A, Jamil M, Cowart LA. Sphingolipids in the Heart: From Cradle to Grave. Front Endocrinol. 2020;11:652. doi: 10.3389/fendo.2020.00652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Borodzicz S, Czarzasta K, Kuch M, et al. Sphingolipids in cardiovascular diseases and metabolic disorders. Lipids Health Dis. 2015;14:55. doi: 10.1186/s12944-015-0053-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang L, Hu Y, An Y, et al. The Changes of Lipidomic Profiles Reveal Therapeutic Effects of Exenatide in Patients With Type 2 Diabetes. Front Endocrinol. 2022;13:677202. doi: 10.3389/fendo.2022.677202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Smith AB, Schill JP, Gordillo R, et al. Ceramides are early responders in metabolic syndrome development in rhesus monkeys. Sci Rep. 2022;12:9960. doi: 10.1038/s41598-022-14083-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Poss AM, Maschek JA, Cox JE, et al. Machine learning reveals serum sphingolipids as cholesterol-independent biomarkers of coronary artery disease. J Clin Invest. 2020;130:1363–76. doi: 10.1172/JCI131838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Hoogendoorn A, Hoedt S, Hartman EMJ, et al. Variation in Coronary Atherosclerosis Severity Related to a Distinct LDL (Low-Density Lipoprotein) Profile Findings From a Familial Hypercholesterolemia Pig Model. Arterioscl Throm Vas. 2019;39:2338–52. doi: 10.1161/ATVBAHA.119.313246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ekroos K, Jänis M, Tarasov K, et al. Lipidomics: a tool for studies of atherosclerosis. Curr Atheroscler Rep. 2010;12:273–81. doi: 10.1007/s11883-010-0110-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hilvo M, Meikle PJ, Pedersen ER, et al. Development and validation of a ceramide- and phospholipid-based cardiovascular risk estimation score for coronary artery disease patients. Eur Heart J. 2020;41:371–80. doi: 10.1093/eurheartj/ehz387. [DOI] [PubMed] [Google Scholar]
- 16.Xie P, Xie J-B, Xiao M-Y, et al. Liver lipidomics analysis reveals the anti-obesity and lipid-lowering effects of gypnosides from heat-processed Gynostemma pentaphyllum in high-fat diet fed mice. Phytomedicine. 2023;115:154834. doi: 10.1016/j.phymed.2023.154834. [DOI] [PubMed] [Google Scholar]
- 17.Liang L, Li D, Zeng R, et al. Long- and very long-chain ceramides are predictors of acute kidney injury in patients with acute coronary syndrome: the PEACP study. Cardiovasc Diabetol. 2023;22:92. doi: 10.1186/s12933-023-01831-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.de Carvalho LP, Tan SH, Ow G-S, et al. Plasma Ceramides as Prognostic Biomarkers and Their Arterial and Myocardial Tissue Correlates in Acute Myocardial Infarction. JACC Basic Transl Sci. 2018;3:163–75. doi: 10.1016/j.jacbts.2017.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Beiser A, D’Agostino RB, Sr, Seshadri S, et al. Computing estimates of incidence, including lifetime risk: Alzheimer’s disease in the Framingham Study. The Practical Incidence Estimators (PIE) macro. Stat Med. 2000;19:1495–522. doi: 10.1002/(sici)1097-0258(20000615/30)19:11/12<1495::aid-sim441>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]
- 20.https://www.stats.gov.cn/sj/pcsj/rkpc/d7c Available.
- 21.Piccoli M, Cirillo F, Ghiroldi A, et al. Sphingolipids and Atherosclerosis: The Dual Role of Ceramide and Sphingosine-1-Phosphate. Antioxidants (Basel) 2023;12:143. doi: 10.3390/antiox12010143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li P-L, Zhang Y. Cross talk between ceramide and redox signaling: implications for endothelial dysfunction and renal disease. Handb Exp Pharmacol. 2013;2013:171–97. doi: 10.1007/978-3-7091-1511-4_9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bismuth J, Lin P, Yao Q, et al. Ceramide: a common pathway for atherosclerosis? Atherosclerosis. 2008;196:497–504. doi: 10.1016/j.atherosclerosis.2007.09.018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Edsfeldt A, Dunér P, Ståhlman M, et al. Sphingolipids Contribute to Human Atherosclerotic Plaque Inflammation. Arterioscler Thromb Vasc Biol. 2016;36:1132–40. doi: 10.1161/ATVBAHA.116.305675. [DOI] [PubMed] [Google Scholar]
- 25.Anroedh S, Hilvo M, Akkerhuis KM, et al. Plasma concentrations of molecular lipid species predict long-term clinical outcome in coronary artery disease patients. J Lipid Res. 2018;59:1729–37. doi: 10.1194/jlr.P081281. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Meeusen JW, Donato LJ, Bryant SC, et al. Plasma Ceramides. Arterioscler Thromb Vasc Biol. 2018;38:1933–9. doi: 10.1161/ATVBAHA.118.311199. [DOI] [PubMed] [Google Scholar]
- 27.Kauhanen D, Sysi-Aho M, Koistinen KM, et al. Development and validation of a high-throughput LC-MS/MS assay for routine measurement of molecular ceramides. Anal Bioanal Chem. 2016;408:3475–83. doi: 10.1007/s00216-016-9425-z. [DOI] [PubMed] [Google Scholar]
- 28.Havulinna AS, Sysi-Aho M, Hilvo M, et al. Circulating Ceramides Predict Cardiovascular Outcomes in the Population-Based FINRISK 2002 Cohort. Arterioscler Thromb Vasc Biol. 2016;36:2424–30. doi: 10.1161/ATVBAHA.116.307497. [DOI] [PubMed] [Google Scholar]
- 29.Hilvo M, Wallentin L, Ghukasyan Lakic T, et al. Prediction of Residual Risk by Ceramide-Phospholipid Score in Patients With Stable Coronary Heart Disease on Optimal Medical Therapy. J Am Heart Assoc. 2020;9:e015258. doi: 10.1161/JAHA.119.015258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Zhao Z, Hou X, Li S, et al. Progress, Challenges, and Opportunities of Geriatric Medicine in China. Aging Dis. 2022;13:1314–6. doi: 10.14336/AD.2022.0128. [DOI] [PMC free article] [PubMed] [Google Scholar]