Abstract
Abstract
Objectives
The atherogenic index of plasma (AIP) has been reported as a biomarker for cardiovascular disease risks and clinical outcomes. However, few studies have investigated the relationship between the AIP and major adverse cardiovascular events (MACEs) after percutaneous coronary intervention (PCI) in patients with ST-segment elevation myocardial infarction (STEMI).
Design
Prospective observational study.
Setting
Data were collected from consecutive patients with STEMI who received PCI at Fuyang People’s Hospital from January 2023 to March 2024.
Patients
A total of 334 patients with STEMI who underwent PCI. The adjusted AIP (aAIP) was calculated using the following formula: aAIP=log ((triglyceride/high-density lipoprotein cholesterol × 100). The patient population was divided into four groups based on the aAIP quartiles.
Primary and secondary outcome measures
The primary outcome was MACEs, and the secondary outcomes included all-cause mortality, all-cause readmission and unplanned readmission as individual endpoints.
Results
Among 334 eligible patients, 68 (20.36%) experienced MACEs during a median follow-up of 8.70 months. After adjusting for confounders, continuous aAIP was positively correlated with MACEs in all three models. Patients in Q3 and Q4 had significantly higher MACE risks than Q1 (p<0.001 for each model), but no difference was observed between Q1 and Q2 (p>0.05 for each model). Compared with Q1, Q4 had significantly increased all-cause mortality risk (Model 2: HR=12.72, 95% CI 1.39 to 76.44 and Model 3: HR=16.18, 95% CI 1.66 to 117.50). All-cause readmission risk was also higher in Q3 and Q4 (Model 3: Q3: HR=2.242, 95% CI 1.043 to 4.818, p=0.039 and Q4: HR=5.378, 95% CI 2.557 to 11.314, p<0.001). Kaplan–Meier curves showed that higher aAIP was associated with increased risks of MACEs, all-cause readmission and unplanned readmission (all log-rank p<0.0001), but not all-cause mortality (log-rank p=0.1534). No significant interactions between subgroups and the aAIP were observed (p>0.05 for interaction).
Conclusions
Higher aAIP was significantly associated with increased risks of MACEs, all-cause readmission and unplanned readmission.
Keywords: Coronary intervention; Myocardial infarction; Death, Sudden, Cardiac
Strengths and limitations of this study.
The adjusted atherogenic index of plasma (aAIP) was used instead of atherogenic index of plasma to avoid negative values and ensure a normal distribution.
Potential confounders were adjusted in multivariable Cox regression models to enhance the robustness of the association.
Key lipid parameters required for aAIP calculation were available for all enrolled patients, minimising missing data bias.
The parameters used to calculate aAIP were collected during the hospital stay, and changes in aAIP were not monitored during follow-up.
Introduction
ST-segment elevation myocardial infarction (STEMI) is a severe type of heart attack that occurs when an atherosclerotic plaque ruptures, causing a sudden decrease in blood flow to the heart. This condition leads to the high rates of disability and mortality. Over the past two decades, reperfusion therapies, such as percutaneous coronary intervention (PCI), thrombolytic therapy and coronary artery bypass grafting, have greatly improved the survival rate of patients with STEMI.1 2 Although these interventions greatly reduce the rates of death and complications, a subset of patients remains at risk for major adverse cardiovascular events (MACEs) post-PCI, including recurrent infarction, repeat revascularisation and cardiac death.3 A meta-analysis of ten randomised trials reported the MACE rates of 10.4% (single-vessel disease) and 16.6% (multivessel disease) in patients with STEMI.4 Similarly, a multicentre retrospective cohort study found no significant reduction in 3-year all-cause mortality (15.7% vs 15.5%; HR=0.92, 95% CI 0.83 to 1.03, p=0.14) between 2005 and 2007 and 2011 and 2012.5 The China PEACE study also noted stable in-hospital mortality from 2001 to 2011 (OR 0.84, 95% CI 0.62 to 1.12, p for trend=0.06).6 These findings highlight the unmet need for early risk stratification tools to identify patients with high-risk STEMI post-PCI.
Several prognostic indices have been proposed for MACE prediction in STEMI, such as the atherogenic index of plasma (AIP), triglyceride-glucose index (TyG),7 glycaemic dispersion index,8 motion score index9 and haemoglobin (HGB), albumin, lymphocyte and platelet score index.10 Among those, the AIP, a logarithmically transformed ratio of triglyceride (TG) to high-density lipoprotein cholesterol (HDL-C), has emerged as a promising biomarker for lipid metabolism dysregulation and coronary artery disease (CAD) severity.11 A meta-analysis linked elevated AIP to a 63% higher MACE risk in patients with CAD (Risk Ratio (RR)=1.63, 95% CI 1.44 to 1.85, p<0.001).12 Furthermore, the AIP’s prognostic value extends to metabolic syndrome,13 non-diabetic CAD,14 type 2 diabetes,15 hypertension16 and the general population,17 with higher AIP levels predicting stroke, myocardial infarction and MACEs. However, few studies have examined the relationship between AIP levels and MACEs in patients with STEMI postdischarge.
This study aimed to explore the association between AIP and MACEs following PCI in patients with STEMI, with all-cause mortality, all-cause readmission and unplanned readmission as secondary endpoints. To mitigate negative values and normalise the distribution, we used an adjusted AIP (aAIP). By elucidating this association, our findings may enhance post-PCI risk stratification and improve cardiovascular outcomes in STEMI.
Methods
Study design and population
This study was a prospective observational study. Patients with STEMI who underwent PCI at Fuyang People’s Hospital were enrolled from January 2023 to March 2024. The permission to conduct the study was approved by the Ethics Committee of Fuyang People’s Hospital (No. 202311).
Patients were included based on the following criteria: (1) patients with STEMI underwent PCI and (2) age≥18 years. Patients were excluded as following criteria: (1) patients with blood diseases or malignant tumours; (2) patients with unmeasured TG or HDL-C; (3) patients who had previously undergone coronary stenting; (4) patients with severe lipid profile metabolic dysfunction,14 indicated by serum bilirubin levels of 5 mg/dL (85 µmol/L) or higher, an international normalised ratio of 1.5 or greater, with prothrombin activity below 40%, or a glomerular filtration rate (GFR) less than 30 mL/min/1.73 m2; (5) patients who did not undergo PCI during hospitalisation or (6) patients with incomplete follow-up.
Sample size
The sample size was calculated using PASS software (V.15.0). Based on the previously reported 1-year MACE rates (overall 9.7%, with AIP quartile groups at 6.3%, 9.9%, 10.3%, and 12.4%, respectively),18 with a two-sided α=0.05, 80% power and accounting for a 5% loss to follow-up, the calculated minimum required sample size was 335 patients.
Data collection
Trained researchers collected clinical data using a standardised questionnaire to ensure consistency. Researchers recorded the patients' demographic characteristics and underlying diseases, which included sex, age, height, weight, smoking status, alcohol consumption, hypertension, type 2 diabetes and stroke. Venous blood samples were collected on the second day of hospital admission after an overnight fast. The following biological markers were measured: total cholesterol (TC), TG, low-density lipoprotein cholesterol (LDL-C), HDL-C, fasting blood glucose, B-type natriuretic peptide, creatine kinase, creatine kinase myocardial band, cardiac troponin, lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase and serum creatinine levels. The platelet count and HGB level from routine blood tests, along with the left ventricular ejection fraction from echocardiography, were also recorded. The medications administered at discharge were documented, including lipid-lowering agents (such as statins), antiplatelet agents (such as aspirin, clopidogrel and ticagrelor), ACE inhibitors, angiotensin II receptor blockers, angiotensin receptor-neprilysin inhibitors and beta-blockers. Comorbidities in this study were defined via International Classification of Diseases (ICD)−10 codes. The ICD-10 codes I10.0, E11.9 and I63 correspond to hypertension, diabetes mellitus and stroke, respectively. Current smoking was defined as smoking one cigarette per day for at least 6 months, whereas current drinking was defined as consuming alcohol at least one time a week for the same duration. Body mass index (BMI) was calculated as weight (kg) divided by height squared (m2). Creatinine clearance (Ccr) was used to estimate the GFR. Ccr was calculated according to the Cockcroft–Gault formula .
Definition and grouping of aAIP
The AIP was calculated using the formula AIP=log (TG/HDL-C).19 To avoid negative values and ensure a normal distribution, we adjusted the AIP (aAIP) by multiplying the TG/HDL-C ratio by 100 before taking the logarithm, resulting in the formula: aAIP=log ((TG/HDL-C) ×100). Based on the calculated aAIP values, the patients were divided into four quartile groups: Q1 (aAIP<1.97), Q2 (1.97≤aAIP<2.18), Q3 (2.18≤aAIP<2.42) and Q4 (aAIP≥2.42).
Endpoints and follow-up
Follow-up for the study was conducted through telephone calls or electronic medical records every 3 months. Relevant medical information was collected during each follow-up. The primary outcome of the study was the occurrence of MACEs. The secondary outcomes included all-cause mortality, all-cause readmission and unplanned readmission as individual endpoints. MACEs were defined as non-fatal myocardial infarction, ischaemic stroke, target vessel revascularisation, all-cause death and unplanned readmission for cardiac causes. Non-fatal myocardial infarction was defined as type I myocardial infarction occurring after PCI. Target vessel revascularisation was defined as angina or ischaemia related to the target vessel necessitating repeat PCI or coronary artery bypass graft. All-cause mortality was defined as death from any cause occurring after the index STEMI discharge. All-cause readmission was defined as any unplanned or planned hospital readmission occurring after discharge, regardless of primary diagnosis. Emergency department visits without subsequent hospitalisation were excluded. Unplanned readmission for cardiac causes was defined as readmission for unstable angina pectoris, acute exacerbation of chronic heart failure, malignant arrhythmia, etc. Endpoint outcomes were collected either at the end of follow-up in September 2024 or at 12 months postenrolment, whichever came first.
Statistical analysis
Data analysis and figure creation were carried out via SPSS (V.26), GraphPad Prism (V.8.3.0) and R (V.4.4.0). Normally distributed continuous data are expressed as the mean±SD and were analysed for statistical significance via one-way Analysis of Variance (ANOVA). Non-normally distributed continuous data are expressed as medians and IQRs and were analysed by the Kruskal–Wallis test. Categorical variables were compared using the Pearson χ2 test. Baseline characteristics were stratified by aAIP values. Patients were divided into four groups according to the quartile level of the aAIP, with the lowest quartile as the reference class. In addition to its use as a categorical variable, aAIP was also analysed as a continuous variable to enhance the reliability of the results. Multivariate Cox regression was used to evaluate the relationships between aAIP and various outcomes, including MACEs, all-cause mortality, overall readmission and unplanned readmission. Trend tests were calculated by transforming aAIP quartiles into continuous variables. The results from the Cox regression analysis are presented as HRs and 95% CIs. Model 1 represented the unadjusted data. In Model 2, the data were adjusted for age, sex and BMI. In Model 3, the results were adjusted for age, sex, BMI, Ccr, current drinking, hypertension, type 2 diabetes and LDL-C. In the subgroup analysis, possible modifications of the association between aAIP and clinical outcomes were assessed for variables, including sex (female vs male), age (<60 years vs ≥60 years), Ccr (<80 mL/min vs ≥80 mL/min), current drinking (yes vs no), hypertension (yes vs no), type 2 diabetes (yes vs no), BMI (<24 kg/m2 vs ≥24 kg/m2) and LDL-C (<3.4 mmol/L vs ≥3.4 mmol/L). Multivariate Cox regression models were used in the analysis of interaction. The mean imputation method was applied to impute the missing data of continuous variables. A two-sided p value <0.05 was considered to indicate statistical significance.
Results
Characteristics of the study patients
Figure 1 illustrates a schematic representation of the selection process. Among 483 patients with STEMI hospitalised at Fuyang People’s Hospital between January 2023 and March 2024, certain patients were excluded primarily due to the absence of PCI during hospitalisation (n=41), severe lipid profile metabolic dysfunction (n=27), lack of available data on TG and HDL-C (n=25) and a history of coronary stenting (n=23). Furthermore, 11 patients were lost to follow-up. Consequently, 334 eligible patients were included and categorised into four groups based on the quartile level of the aAIP, with 83 individuals in the Q1 group, 84 in the Q2 group, 84 in the Q3 group and 83 in the Q4 group.
Figure 1. The flowchart of the study participants. aAIP, adjusted atherogenic index of plasma; HDL-C, high-density lipoprotein cholesterol; PCI, percutaneous coronary intervention; TG, triglyceride.
Table 1 presents the baseline characteristics of the patients grouped by aAIP quartile. The average age of the patients at baseline was 62.21±13.30 years, with 253 (75.75%) being male. Higher aAIP groups were younger (p=0.004) and had elevated BMI (Q4: 24.14 kg/m² vs Q1: 22.51 kg/m²; p=0.011). Hypertension (Q4: 65.06% vs Q1: 39.76%; p=0.004) and type 2 diabetes (Q4: 37.35% vs Q1: 10.84%; p<0.001) were more prevalent in higher quartiles. Metabolic dysfunction markers (TC, TG, LDL-C and HDL-C; all p<0.05), current drinking and Ccr also differed significantly across groups.
Table 1. Baseline characteristics of participants categorised by aAIP quartiles.
| Variables | Total (n=334) | aAIP quartiles | P value | |||
|---|---|---|---|---|---|---|
| Q1 (n=83) | Q2 (n=84） | Q3 (n=84) | Q4 (n=83) | |||
| Male, n (%) | 253 (75.75) | 64 (77.11) | 59 (70.24) | 63 (75.00) | 67 (80.72) | 0.455 |
| Age, year | 62.21±13.30 | 64.23±13.24 | 63.43±11.87 | 63.48±12.28 | 57.69±14.82*†‡ | 0.004 |
| BMI, kg/m2 | 23.32±3.48 | 22.51±3.22 | 22.94±3.10 | 23.70±3.11* | 24.14±4.21*† | 0.011 |
| Current smoking, n (%) | 185 (55.39) | 49 (59.04) | 42 (50.00) | 43 (51.19) | 51 (61.44) | 0.352 |
| Current drinking, n (%) | 139 (41.62) | 32 (38.55) | 28 (33.33) | 33 (39.29) | 46 (55.42)† | 0.025 |
| Hypertension, n (%) | 186 (55.69) | 33 (39.76) | 46 (54.76) | 53 (63.09)* | 54 (65.06)* | 0.004 |
| Type 2 diabetes, n (%) | 78 (23.35) | 9 (10.84) | 13 (15.48) | 25 (29.76)*† | 31 (37.35)*† | <0.001 |
| Stroke, n (%) | 43 (12.87) | 11 (13.25) | 10 (11.90) | 12 (14.28) | 10 (12.05) | 0.964 |
| LVEF, (%) | 55.50 (48.00, 61.00) | 55.00 (50.00, 60.50) | 54.50 (47.00, 60.00) | 56.00 (47.00, 61.00) | 56.00 (49.50, 62.25) | 0.828 |
| BNP, pg/mL | 844.00 (282.00, 2118.00) | 523.00 (174.00, 2118.00） | 1118.50 (435.75, 1966.75) | 1043.50 (453.75, 2524.50) | 658.50 (211.00, 2182.50) | 0.060 |
| cTn, ng/mL | 17.49 (3.22, 50.00) | 24.08 (2.33, 50.00) | 13.29 (2.19, 44.34) | 18.85 (4.97, 48.47) | 18.47 (3.21, 50.00) | 0.551 |
| AST, U/L | 5.15 (1.30, 33.72) | 5.60 (1.30, 35.10) | 5.30 (1.75, 31.10) | 8.25 (1.52, 33.58) | 2.80 (1.10, 30.10) | 0.130 |
| LDH, U/L | 343.50 (242.00, 568.00) | 357.60 (245.15, 608.87) | 317.00 (254.70, 576.00) | 356.05 (237.42, 555.02) | 350.60 (211.55, 516.52) | 0.771 |
| CK, U/L | 485.50 (132.30, 1427.55) | 698.60 (143.85, 2094.00) | 447.20 (152.10, 1383.80) | 471.50 (136.50, 1362.60) | 325.00 (115.05, 1281.50) | 0.370 |
| CK-MB, U/L | 28.82 (5.56, 125.29) | 40.82 (6.04, 181.04) | 26.16 (6.48, 97.26) | 30.18 (5.24, 129.10) | 23.97 (5.14, 98.25) | 0.315 |
| FBG, mmol/L | 6.08 (5.03, 7.80) | 5.60 (4.82, 6.96) | 6.25 (5.07, 7.76) | 6.26 (4.87, 7.87) | 6.30 (5.22, 8.47) | 0.064 |
| HGB, g/L | 137.00 (123.00, 150.25) | 137.00 (127.00, 149.50) | 136.00 (122.00, 148.00) | 133.00 (119.75, 149.25) | 140.00 (125.75, 157.25) | 0.132 |
| PLT, ×109 /L | 216.00 (170.50, 259.00) | 201.00 (164.50, 247.00) | 209.00 (159.00, 256.00) | 223.00 (175.00, 251.00) | 224.50 (184.25, 277.00) | 0.127 |
| Ccr, mL/min | 85.42 (65.08, 108.90) | 84.36 (64.79, 100.88) | 78.94 (62.22, 104.08) | 88.15 (61.69, 104.93） | 96.21 (70.43, 121.59)† | 0.047 |
| ALT, U/L | 24.75 (16.52, 37.45) | 23.10 (15.40, 36.30) | 24.70 (16.00, 37.60) | 23.60 (16.30, 36.80) | 28.75 (19.10, 41.45) | 0.115 |
| TC, mmol/L | 4.42 (3.76, 5.17) | 4.20 (3.56, 4.88) | 4.32 (3.74, 5.09) | 4.86 (4.12, 5.53)* | 4.40 (3.66, 5.18) | 0.002 |
| TG, mmol/L | 1.40 (0.97, 1.98) | 0.87 (0.69, 1.07) | 1.31 (1.03, 1.62)* | 1.60 (1.04, 2.32)*† | 2.13 (1.56, 3.35)*†‡ | <0.001 |
| HDL-C, mmol/L | 0.99 (0.85, 1.18) | 1.17 (0.97, 1.34) | 1.01 (0.86, 1.19)* | 0.97 (0.84, 1.14)* | 0.85 (0.74, 1.03)*†‡ | <0.001 |
| LDL-C, mmol/L | 2.54 (2.03, 3.13) | 2.34 (1.75, 2.83) | 2.52 (2.11, 3.07) | 2.42 (1.89, 2.89) | 2.96 (2.45, 3.66)*†‡ | <0.001 |
| Tirofiban, n (%) | 106 (31.74) | 23 (27.71) | 30 (35.71) | 27 (32.14) | 26 (31.32) | 0.742 |
| Number of stents, n (%) | 0.354 | |||||
| 1 | 211 (63.17) | 52 (62.65) | 57 (67.86) | 49 (58.33) | 53 (63.86) | |
| 2 | 90 (26.95) | 18 (21.69) | 22 (26.19) | 27 (32.14) | 23 (27.71) | |
| ≥3 | 33 (9.88) | 13 (15.66) | 5 (5.95) | 8 (9.52) | 7 (8.43) | |
| aAIP | 2.18 (1.97, 2.42) | 1.85 (1.77, 1.93) | 2.05 (2.01, 2.12)* | 2.28 (2.21, 2.35)*† | 2.60 (2.49, 2.75)*†‡ | <0.001 |
| Medication at discharge, n (%) | ||||||
| Aspirin+clopidogrel | 61 (18.26) | 8 (9.64) | 15 (17.86) | 22 (26.19) | 16 (19.28) | 0.052 |
| Aspirin+ticagrelor | 273 (81.74) | 75 (90.36) | 69 (82.14) | 62 (73.81) | 67 (80.72) | 0.052 |
| Statins | 324 (97.00) | 80 (96.38) | 82 (97.62) | 82 (97.62) | 80 (96.38) | 0.932 |
| ACEI/ARB/ARNI | 251 (75.15) | 61 (73.49) | 63 (75.00) | 66 (78.57) | 61 (73.49) | 0.856 |
| β-blocker | 210 (62.87) | 50 (60.24) | 57 (67.86) | 51 (60.71) | 52 (62.65) | 0.727 |
Data were presented as mean±SD, median and IQR, or as n(%).
Compared with Q1, p<0.05.
Compared with Q2, p<0.05.
Compared with Q3, p<0.05.
aAIP, adjusted atherogenic index of plasma; ACEI, ACE inhibitors; ALT, alanine aminotransferase; ARB, angiotensin II receptor blockers; ARNI, angiotensin receptor-neprilysin inhibitors; AST, aspartate aminotransferase; BMI, body mass index; BNP, B-type natriuretic peptide; Ccr, creatinine clearance; CK, creatine kinase; CK-MB, creatine kinase myocardial band; cTn, cardiac troponin; FBG, fasting blood glucose; HDL-C, high-density lipoprotein cholesterol; HGB, haemoglobin; LDH, lactate dehydrogenase; LDL-C, low-density lipoprotein cholesterol; LVEF, left ventricular ejection fraction; PLT, platelet count; TC, total cholesterol; TG, triglyceride; β-blocker, beta-blocker.
Association between aAIP and primary outcome
Over a median follow-up of 8.70 months, 20.36% (n=68) patients experienced MACEs, including unstable angina (8.98%), acute heart failure exacerbation (3.89%), non-fatal MI (3.59%), all-cause death (2.99%), ischaemic stroke (1.50%) and malignant arrhythmia (0.60%). Continuous aAIP was positively correlated with MACEs in Model 1. This correlation remained significant after adjusting for sex, age and BMI in Model 2, and further adjusted for Ccr, current drinking, hypertension, type 2 diabetes and LDL-C in Model 3 (p<0.001 for each model). Quartile analysis revealed significantly higher MACE risks in Q3 and Q4 versus Q1 (all p<0.001), but no Q1–Q2 difference (p>0.05). A dose-dependent trend was observed (p<0.001 for trend). Kaplan–Meier curves (figure 2A) confirmed progressively higher MACE incidence from Q1 to Q4 (log-rank p<0.0001).
Figure 2. Survival analysis for outcome events. Panels present the Kaplan‒Meier curves for (A) MACEs, (B) all-cause death, (C) all-cause readmission and (D) unplanned readmission in 334 participants stratified into four groups according to aAIP levels. aAIP, adjusted atherogenic index of plasma; MACEs, major adverse cardiovascular events.
Associations between aAIP and secondary outcomes
During the follow-up period, the study recorded 79 cases (23.65%) of all-cause readmission and 68 cases (20.36%) of unplanned readmission (table 2). Continuous aAIP correlated with all-cause death, readmission and unplanned readmission (all p<0.05). Q4 had significantly higher all-cause mortality versus Q1 (Model 2: HR=12.72, 95% CI 1.39 to 76.44 and Model 3: HR=16.18, 95% CI 1.66 to 117.50), though Kaplan–Meier curves showed no quartile difference (log-rank p=0.1534; figure 2B). For all-cause readmission, Q3 (HR=2.242, 95% CI 1.043 to 4.818; p=0.039) and Q4 (HR=5.378, 95% CI 2.557 to 11.314; p<0.001) had elevated risks in Model 3, with a significant trend (p<0.001). Unplanned readmission risks were higher in Q3–Q4 across all models (p<0.05), supported by Kaplan–Meier analysis (log-rank p<0.0001; figure 2C,D).
Table 2. HR (95% CI) of MACEs, all-cause death, all-cause readmission and unplanned readmission according to aAIP in the three models.
| Endpoints | Event, n (%) | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | ||
| MACEs | |||||||
| Continuous variable | 68 (20.36) | 8.06 (4.695 to 13.848) |
＜0.001 | 13.507 (7.427 to 24.564) |
<0.001 | 38.169 (18.040 to 80.759) |
＜0.001 |
| aAIP quartile | |||||||
| Q1 | 3 (0.90) | Ref | Ref | Ref | Ref | Ref | Ref |
| Q2 | 2 (0.60) | 0.666 (0.111 to 3.985) |
0.656 | 0.746 (0.124 to 4.474) |
0.748 | 1.089 (0.174 to 6.840) |
0.927 |
| Q3 | 23 (6.89) | 8.837 (2.652 to 29.445) |
＜0.001 | 10.844 (3.231 to 36.393) |
<0.001 | 17.676 (4.837 to 64.591) |
＜0.001 |
| Q4 | 40 (11.98) | 16.923 (5.229 to 54.773） |
＜0.001 | 26.186 (7.965 to 86.094) |
＜0.001 | 47.077 (13.069 to 169.581) |
＜0.001 |
| P for trend | ＜0.001 | ＜0.001 | ＜0.001 | ||||
| All-cause death | |||||||
| Continuous variable | 10 (2.99) | 5.273 (1.186 to 23.444) |
0.029 | 18.355 (3.003 to 62.210) |
0.002 | 36.951 (4.015 to 140.044) |
0.001 |
| aAIP quartile | |||||||
| Q1 | 1 (0.00) | Ref | Ref | Ref | Ref | Ref | Ref |
| Q2 | 1 (0.00) | 0.980 (0.061 to 15.677) |
0.989 | 1.369 (0.084 to 22.261) |
0.825 | 1.365 (0.083 to 22.489) |
0.828 |
| Q3 | 3 (0.90) | 3.219 (0.335 to 30.964) |
0.311 | 4.736 (0.477 to 46.979) |
0.184 | 5.894 (0.551 to 63.061) |
0.142 |
| Q4 | 5 (1.49) | 5.647 (0.658 to 48.433) |
0.114 | 12.720 (1.389 to 76.436) |
0.024 | 16.181 (1.662 to 117.495) |
0.016 |
| P for trend | 0.037 | 0.007 | 0.005 | ||||
| All-cause readmission | |||||||
| Continuous variable | 79 (23.65) | 3.745 (2.158 to 6.500) |
<0.001 | 4.220 (2.346 to 7.593) |
<0.001 | 4.977 (2.621 to 9.454) |
<0.001 |
| aAIP quartile | |||||||
| Q1 | 12 (3.59) | Ref | Ref | Ref | Ref | Ref | Ref |
| Q2 | 11 (3.29) | 0.912 (0.402 to 2.067) |
0.825 | 0.913 (0.402 to 2.075) |
0.828 | 1.025 (0.439 to 2.393) |
0.955 |
| Q3 | 20 (5.99) | 1.919 (0.938 to 3.928) |
0.074 | 1.973 (0.957 to 4.069) |
0.066 | 2.242 (1.043 to 4.818) |
0.039 |
| Q4 | 36 (10.78) | 3.774 (1.960 to 7.267) |
<0.001 | 4.461 (2.264 to 8.789) |
<0.001 | 5.378 (2.557 to 11.314) |
<0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | ||||
| Unplanned readmission | |||||||
| Continuous variable | 68 (20.36) | 4.737 (2.669 to 8.406) |
<0.001 | 5.555 (3.006 to 10.264) |
<0.001 | 6.646 (3.399 to 12.994) |
<0.001 |
| aAIP quartile | |||||||
| Q1 | 9 (2.69) | Ref | Ref | Ref | Ref | Ref | Ref |
| Q2 | 8 (2.39) | 0.888 (0.343 to 2.303) |
0.807 | 0.895 (0.344 to 2.327) |
0.820 | 1.081 (0.402 to 2.907) |
0.877 |
| Q3 | 18 (5.39) | 2.329 (1.045 to 5.186) |
0.039 | 2.439 (1.085 to 5.480) |
0.031 | 2.945 (1.246 to 6.960) |
0.014 |
| Q4 | 33 (9.88) | 4.732 (2.260 to 9.910) |
<0.001 | 5.706 (2.660 to 12.238) |
<0.001 | 7.281 (3.146 to 16.848) |
<0.001 |
| P for trend | <0.001 | <0.001 | <0.001 | ||||
Model 1 represented the unadjusted data.
Model 2 adjusted for age, sex and BMI.
Model 3 adjusted for age, sex, BMI, Ccr, current drinking, hypertension, type 2 diabetes and LDL-C.
aAIP, adjusted atherogenic index of plasma; BMI, body mass index; Ccr, creatinine clearance; LDL-C, low-density lipoprotein cholesterol; MACE, major adverse cardiovascular event.
Subgroup analysis
Subgroup analysis was conducted to explore consistency among the significant subgroups. Figure 3 presented the results of the modification effects between aAIP and MACEs in various subgroups. Continuous aAIP was positively correlated with MACEs in all subgroups. No significant interaction effects were observed (all p for interaction >0.05).
Figure 3. Subgroup and interaction analyses of the associations between aAIP and MACEs. Model adjusted for sex, age, Ccr, current drinking, hypertension, type 2 diabetes, BMI and LDL-C. aAIP, adjusted atherogenic index of plasma; BMI, body mass index; Ccr, creatinine clearance rate; LDL-C, low-density lipoprotein cholesterol; MACEs, major adverse cardiovascular events.
Discussion
In this observational study, we found a significant association between elevated aAIP and increased risks of MACEs, all-cause readmission and unplanned readmission in patients with STEMI post-PCI, independent of traditional cardiovascular risk factors. These findings highlight the potential clinical utility of aAIP as a prognostic tool for risk stratification in this high-risk population.
STEMI is a critical cardiovascular condition characterised by sudden blockage of blood flow in the coronary arteries, leading to myocardial injury. It remains a leading cause of morbidity and mortality worldwide, significantly affecting the patient quality of life and healthcare systems.20 Despite advancements in PCI therapeutic approaches, challenges persist, including the risk of MACEs and the need for improved prognostic markers to improve patient outcomes.2 21 The AIP, defined as the logarithm of the TG to HDL-C ratio, serves as a significant indicator of lipid metabolism disturbances and cardiovascular risk. Several studies have demonstrated that the AIP was associated with various cardiovascular diseases (CVDs) and conditions, acting as a potential prognostic marker for MACEs. Abdu et al22 reported that the AIP was notably associated with the incidence of MACEs in myocardial infarction patients and non-obstructive coronary artery patients. Yao et al23 reported that, for each additional unit of the AIP, the risk of MACEs increased by 19% in coronary artery calcification patients. They suggested that AIP was a readily available and low-cost strategy for identifying MACEs and mediating the association between coronary artery calcification and MACEs. However, Karakayali et al24 found that although both AIP and TyG index were significantly higher in patients with non-obstructive CADs compared with the control group, the TyG index provided a stronger prediction than AIP. In addition, other studies have indicated that AIP was also a useful prognostic tool in non-diabetic CAD,14 chronic coronary syndrome25 and type 2 diabetes mellitus.15 However, the application of the AIP in predicting MACEs after PCI in patients with STEMI remains limited in the literature. In previous reports, negative values were frequently observed in AIP calculations, primarily because of the calculation formula.12,17 However, the negative values of AIP posed challenges in data processing and statistical analysis, which complicated the accurate interpretation of the results. In this study, aAIP was used instead of AIP to avoid negative values and ensure a normal distribution of data. This study observed that a higher aAIP was significantly correlated with an increased risk of MACEs, as well as a notable rise in all-cause and unplanned readmission rates. These findings indicated that the aAIP might serve as a valuable tool for risk stratification in clinical practice, assisting in the identification of patients with STEMI at increased risk of MACEs.
The precise mechanisms linking aAIP to MACEs in patients with STEMI after PCI remain unclear. One possible explanation is that atherogenicity might be a significant factor, as aAIP is derived from the formulae for TG and HDL-C. When the plasma TG level was relatively high, LDL-C particle phenotype tended to be smaller, denser, more easily oxidised and more prone to enter the subintima, thereby increasing their atherogenic potential.26,28 Our results also revealed that a higher AIP was associated with traditional cardiovascular risk factors, including hypertension, type 2 diabetes and metabolic disorders. This strengthens its role in identifying high-risk patients who might need more intervention measures. This finding is consistent with previous findings that highlighted the importance of a comprehensive risk assessment in the effective management of CVD.29
While a Lithuanian cohort study found AIP significantly predicted CVD mortality in men and all-cause mortality in women,30 our study showed conflicting results. Although adjusted models revealed higher all-cause mortality risk in Q4 versus Q1 (Models 2 and 3), Kaplan–Meier analysis demonstrated no significant cumulative mortality difference (log-rank p=0.1534). This aligns with two large cohort studies showing no significant AIP–mortality association after full adjustment,31 32 despite trending HRs. The discrepancy may reflect our shorter follow-up duration (8.7 months vs 10 years31), suggesting that extended observation may be needed to detect mortality effects. These mixed findings highlight the importance of considering study design and follow-up duration when interpreting AIP’s prognostic value for mortality outcomes.
Our study indeed demonstrated a significant association between higher aAIP and an elevated risk of MACEs, as well as all-cause and unplanned readmissions. For patients with STEMI with high aAIP levels, particularly those in Q3 and Q4, we propose the importance of a more personalised and comprehensive management strategy. Specifically, patients with higher aAIP levels may benefit from more intensive cardiovascular risk factor management, including aggressive lipid-lowering therapy, tighter blood pressure control and enhanced lifestyle interventions. Additionally, closer monitoring and follow-up may be warranted to detect and manage potential complications early. However, these recommendations require validation in prospective intervention trials specifically targeting aAIP reduction. Future studies should investigate whether aAIP-guided therapy can improve outcomes in this high-risk population.
Although AIP has been studied in CVD, its prognostic value in patients with STEMI undergoing PCI remains unclear. Our study focuses on this high-risk cohort, which has distinct pathophysiological characteristics compared with other cardiovascular populations. We provide insights that a higher aAIP is significantly associated with increased risks of MACEs, all-cause readmission and unplanned readmission. These findings are directly relevant to the management and prognosis of patients with STEMI treated with PCI. To improve risk stratification, we introduced an aAIP formula that addresses a key limitation of the conventional AIP calculation. The traditional formula can yield negative values, reducing its utility in risk assessment. In contrast, the aAIP formula eliminates this issue, enabling more robust and clinically meaningful risk stratification. This adjustment enhances the applicability and reliability of AIP as a prognostic marker in PCI-treated patients. In addition to traditional MACEs, our study identifies the association between aAIP and unplanned readmissions. Unplanned readmissions reflect persistent health challenges and increased healthcare utilisation, carrying significant clinical and economic implications. Our findings broaden the understanding of aAIP’s prognostic impact and its relevance to post-PCI outcomes.
This study has several limitations. First, this was a single-centre observational study with a relatively small sample size, which may limit the generalisability of the findings. Second, the adjusted covariates were only collected at baseline, and they may have changed over time. Additionally, residual confounding factors cannot be entirely excluded. These factors may potentially influence the association of aAIP with MACEs in patients with STEMI after PCI. Third, the parameters used to calculate aAIP were collected during the hospital stay, and changes in aAIP were not monitored during follow-up. Some factors, such as dietary patterns and environmental factors, may affect the value of aAIP, potentially leading to bias. Hence, further studies are needed to evaluate the application of aAIP among patients with STEMI after PCI.
Conclusion
This study demonstrated that elevated aAIP levels were independently associated with higher risks of MACEs, all-cause mortality and unplanned readmission in patients with STEMI after PCI. These findings highlight the potential of aAIP as a novel prognostic biomarker for post-PCI risk stratification. By identifying high-risk patients during the acute phase, aAIP measurement may facilitate timely secondary prevention strategies to improve clinical outcomes. Further studies are needed to validate its utility in prehospital settings and to explore targeted interventions for patients with elevated aAIP.
Acknowledgements
We are indebted to all the members who contributed to this work.
Footnotes
Funding: This study was supported by the Anhui Provincial Health Research Project (Grant No. AHWJ2023A30162), the Fuyang City Key Research and Development Plan for Clinical Medical Research Translation Special Project (Grant No. FYZDYF2023LCYX012) and the Bengbu Medical College Research Project (Grant No. 2022byzd166).
Prepublication history for this paper is available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2024-096065).
Data availability free text: All data generated or analysed during this study are included in this article, along with references to data from cited published studies. The database is not publicly available. Further inquiries can be directed to the corresponding author.
Patient consent for publication: Consent obtained directly from patient(s).
Ethics approval: This study involves human participants and was approved by Ethics Committee of Fuyang People’s Hospital (No. 202311). Participants gave informed consent to participate in the study before taking part.
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
Data are available on reasonable request.
References
- 1.Levi F, Chatenoud L, Bertuccio P, et al. Mortality from cardiovascular and cerebrovascular diseases in Europe and other areas of the world: an update. Eur J Cardiovasc Prev Rehabil. 2009;16:333–50. doi: 10.1097/HJR.0b013e328325d67d. [DOI] [PubMed] [Google Scholar]
- 2.Wu Z-F, Su W-T, Chen S, et al. PTH Predicts the in-Hospital MACE After Primary Percutaneous Coronary Intervention for Acute ST-Segment Elevation Myocardial Infarction. Ther Clin Risk Manag. 2023;19:699–712. doi: 10.2147/TCRM.S420335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Byrne RA, Rossello X, Coughlan JJ, et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44:3720–826. doi: 10.1093/eurheartj/ehad191. [DOI] [PubMed] [Google Scholar]
- 4.Bajraktari G, Bytyçi I, Henein MY, et al. Complete revascularization for patients with multivessel coronary artery disease and ST-segment elevation myocardial infarction after the COMPLETE trial: A meta-analysis of randomized controlled trials. Int J Cardiol Heart Vasc. 2020;29:100549. doi: 10.1016/j.ijcha.2020.100549. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Takeji Y, Shiomi H, Morimoto T, et al. Changes in demographics, clinical practices and long-term outcomes of patients with ST segment-elevation myocardial infarction who underwent coronary revascularisation in the past two decades: cohort study. BMJ Open. 2021;11:e043683. doi: 10.1136/bmjopen-2020-043683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Li J, Li X, Wang Q, et al. ST-segment elevation myocardial infarction in China from 2001 to 2011 (the China PEACE-Retrospective Acute Myocardial Infarction Study): a retrospective analysis of hospital data. Lancet. 2015;385:441–51. doi: 10.1016/S0140-6736(14)60921-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wang W, Yang J, Wang K, et al. Association between the triglyceride-glucose index and in-hospital major adverse cardiovascular events in patients with acute coronary syndrome: results from the Improving Care for Cardiovascular Disease in China (CCC)-Acute Coronary Syndrome project. Cardiovasc Diabetol. 2024;23:170. doi: 10.1186/s12933-024-02270-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shi R, Xu W, Feng L, et al. Value of Glycemic Dispersion Index in Predicting Major Adverse Cardiovascular Events in Diabetic Patients with Concomitant Acute Coronary Syndrome. Diabetes Metab Syndr Obes. 2024;17:3433–45. doi: 10.2147/DMSO.S469436. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kong F, Xiang L, Wu Y, et al. Evaluation of the Prognostic Role of the Wall Motion Score Index and the SYNTAX Score II in Patients with Acute Coronary Syndrome Following Percutaneous Coronary Intervention by Evaluation of Major Adverse Cardiovascular Events at 12-Month Follow-Up. Med Sci Monit. 2021;27:e932652. doi: 10.12659/MSM.932652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Karakayali M, Omar T, Artac I, et al. The prognostic value of HALP score in predicting in-hospital mortality in patients with ST-elevation myocardial infarction undergoing primary percutaneous coronary intervention. Coron Artery Dis. 2023;34:483–8. doi: 10.1097/MCA.0000000000001271. [DOI] [PubMed] [Google Scholar]
- 11.Huang H, Yu X, Li L, et al. Atherogenic index of plasma is related to coronary atherosclerotic disease in elderly individuals: a cross-sectional study. Lipids Health Dis. 2021;20:68. doi: 10.1186/s12944-021-01496-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rabiee Rad M, Ghasempour Dabaghi G, Darouei B, et al. The association of atherogenic index of plasma with cardiovascular outcomes in patients with coronary artery disease: A systematic review and meta-analysis. Cardiovasc Diabetol. 2024;23:119. doi: 10.1186/s12933-024-02198-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Krishnan M, Faculty of Allied Health Sciences, Chettinad Hospital and Research Institute, Chettinad Academy of Research and Education, Kelambakkam-603103, Tamilnadu, India Atherogenic index of plasma, lipid accumulation and visceral adiposity in metabolic syndrome patients. Bioinformation. 2022;18:1109–13. doi: 10.6026/973206300181109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Zheng Y, Li C, Yang J, et al. Atherogenic index of plasma for non-diabetic, coronary artery disease patients after percutaneous coronary intervention: a prospective study of the long-term outcomes in China. Cardiovasc Diabetol. 2022;21:29. doi: 10.1186/s12933-022-01459-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Fu L, Zhou Y, Sun J, et al. Atherogenic index of plasma is associated with major adverse cardiovascular events in patients with type 2 diabetes mellitus. Cardiovasc Diabetol. 2021;20:201. doi: 10.1186/s12933-021-01393-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Karayiğit O, Dolu AK, Çelik MC, et al. Relationship between the Atherogenic Index of Plasma and Nondipping Circadian Pattern in Hypertensive Patients. Med Princ Pract. 2023;32:26–32. doi: 10.1159/000527662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Zhang Y, Chen S, Tian X, et al. Association between cumulative atherogenic index of plasma exposure and risk of myocardial infarction in the general population. Cardiovasc Diabetol. 2023;22:210. doi: 10.1186/s12933-023-01936-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Liu Z, Zhang L, Wang L, et al. The predictive value of cumulative atherogenic index of plasma (AIP) for cardiovascular outcomes: a prospective community-based cohort study. Cardiovasc Diabetol. 2024;23:264. doi: 10.1186/s12933-024-02350-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fernández-Macías JC, Ochoa-Martínez AC, Varela-Silva JA, et al. Atherogenic Index of Plasma: Novel Predictive Biomarker for Cardiovascular Illnesses. Arch Med Res. 2019;50:285–94. doi: 10.1016/j.arcmed.2019.08.009. [DOI] [PubMed] [Google Scholar]
- 20.Butcher SC, Lustosa RP, Abou R, et al. Prognostic implications of left ventricular myocardial work index in patients with ST-segment elevation myocardial infarction and reduced left ventricular ejection fraction. Eur Heart J Cardiovasc Imaging. 2022;23:699–707. doi: 10.1093/ehjci/jeab096. [DOI] [PubMed] [Google Scholar]
- 21.Karakayali M, Ogun M, Artac I, et al. Serum malondialdehyde levels at admission as a predictor of inhospital mortality in patients with acute coronary syndrome. Coron Artery Dis. 2025;36:211–7. doi: 10.1097/MCA.0000000000001469. [DOI] [PubMed] [Google Scholar]
- 22.Abdu FA, Alifu J, Mohammed A-Q, et al. The correlation of atherogenic index of plasma with non-obstructive CAD and unfavorable prognosis among patients diagnosed with MINOCA. Eur J Intern Med. 2024;125:111–9. doi: 10.1016/j.ejim.2024.03.024. [DOI] [PubMed] [Google Scholar]
- 23.Yao H, Feng G, Liu Y, et al. Coronary artery calcification burden, atherogenic index of plasma, and risk of adverse cardiovascular events in the general population: evidence from a mediation analysis. Lipids Health Dis. 2024;23:258. doi: 10.1186/s12944-024-02255-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Karakayalı M, Altunova M, Yakisan T, et al. Relationship Between Nonobstructive Coronary Arteries and Metabolic Parameters. Kafkas J Med Sci . 2024;14:138–43. doi: 10.5505/kjms.2024.86300. [DOI] [Google Scholar]
- 25.Alifu J, Xiang L, Zhang W, et al. Association between the atherogenic index of plasma and adverse long-term prognosis in patients diagnosed with chronic coronary syndrome. Cardiovasc Diabetol. 2023;22:255. doi: 10.1186/s12933-023-01989-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Wang Y, Wang S, Sun S, et al. The predictive value of atherogenic index of plasma for cardiovascular outcomes in patients with acute coronary syndrome undergoing percutaneous coronary intervention with LDL-C below 1.8mmol/L. Cardiovasc Diabetol. 2023;22:150. doi: 10.1186/s12933-023-01888-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Quispe R, Manalac RJ, Faridi KF, et al. Relationship of the triglyceride to high-density lipoprotein cholesterol (TG/HDL-C) ratio to the remainder of the lipid profile: The Very Large Database of Lipids-4 (VLDL-4) study. Atherosclerosis. 2015;242:243–50. doi: 10.1016/j.atherosclerosis.2015.06.057. [DOI] [PubMed] [Google Scholar]
- 28.Guérin M, Le Goff W, Lassel TS, et al. Atherogenic role of elevated CE transfer from HDL to VLDL(1) and dense LDL in type 2 diabetes: impact of the degree of triglyceridemia. Arterioscler Thromb Vasc Biol. 2001;21:282–8. doi: 10.1161/01.atv.21.2.282. [DOI] [PubMed] [Google Scholar]
- 29.Islam MS, Islam MN, Kundu SK, et al. Serum Albumin Level and In-Hospital Outcome of Patients with First Attack Acute Myocardial Infarction. Mymensingh Med J. 2019;28:744–51. [PubMed] [Google Scholar]
- 30.Tamosiunas A, Luksiene D, Kranciukaite-Butylkiniene D, et al. Predictive importance of the visceral adiposity index and atherogenic index of plasma of all-cause and cardiovascular disease mortality in middle-aged and elderly Lithuanian population. Front Public Health. 2023;11:1150563. doi: 10.3389/fpubh.2023.1150563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.You F-F, Gao J, Gao Y-N, et al. Association between atherogenic index of plasma and all-cause mortality and specific-mortality: a nationwide population‑based cohort study. Cardiovasc Diabetol. 2024;23:276. doi: 10.1186/s12933-024-02370-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Kim SH, Cho YK, Kim Y-J, et al. Association of the atherogenic index of plasma with cardiovascular risk beyond the traditional risk factors: a nationwide population-based cohort study. Cardiovasc Diabetol. 2022;21:81. doi: 10.1186/s12933-022-01522-8. [DOI] [PMC free article] [PubMed] [Google Scholar]