Abstract
Background
The platelet-to-albumin ratio (PAR) has shown to be linked with cardiovascular disorders. However, the specific association between the admission PAR and one-year outcomes in patients with acute coronary syndrome (ACS) remains to be fully characterized. This study investigated the linkage between PAR levels and the occurrence of one-year all-cause mortality and major adverse cardiovascular and cerebrovascular events (MACCEs) in the ACS population.
Patients and methods
This retrospective cohort study enrolled patients diagnosed with ACS between January 2022 and December 2023. The primary endpoint was one-year all-cause mortality, and the secondary endpoint was MACCE, defined as a composite of all-cause mortality, non-fatal myocardial infarction, ischemic stroke, heart failure rehospitalization, target vessel revascularization or in-stent thrombosis, and malignant arrhythmia. To evaluate the relationship between PAR and these clinical events, we employed multivariate Cox proportional hazards models, restricted cubic splines (RCSs) and two-piecewise linear regression to identify potential non-linear associations and specific inflection points.
Results
A total of 1326 participants (median age: 67 years) were followed for one year, during which 137 (10.3%) deaths and 280 (21.1%) MACCE events occurred. The one-year all-cause mortality rates across the PAR quartiles (Q1–Q4) were 11.6%, 6.1%, 7.0% and 16.7%, respectively, with a similar trend observed for MACCE (22.1%, 15.2%, 18.7% and 28.5%). RCS analysis revealed significant nonlinear associations between the PAR and both clinical endpoints (all p for nonlinear < 0.05). Two-piecewise linear regression analysis identified specific inflection points at 5.28 (95% confidence interval [CI]: 4.86–9.86) for one-year all-cause mortality and 5.29 (95% CI: 4.37–6.58) for MACCE.
Conclusions
The PAR demonstrates a significant nonlinear relationship with one-year all-cause mortality and MACCE in patients with ACS.
Keywords: Inflammation, platelet-to-albumin ratio, all-cause mortality, major adverse cardiovascular and cerebrovascular events, acute coronary syndrome
Key messages
In patients with acute coronary syndrome, the platelet to albumin ratio is significantly associated with one-year all-cause mortality and major adverse cardiovascular and cerebrovascular events in a non-linear manner.
GRAPHICAL ABSTRACT
Introduction
Acute coronary syndrome (ACS), which encompasses unstable angina (UA), non-ST-segment elevation myocardial infarction (NSTEMI) and ST-segment elevation myocardial infarction (STEMI), remains a leading contributor to global morbidity and mortality [1–3]. Despite significant advancements in early diagnosis, reperfusion strategies and medical therapies, the risk of adverse cardiovascular events and all-cause mortality remains substantial, particularly within the first year following the index event [4,5]. Identifying readily accessible and cost-effective haematological parameters that accurately reflect patient risk stratification is essential for refining management strategies.
The underlying pathophysiology of ACS involves a complex interaction between systemic inflammation and hemodynamic instability, both of which profoundly influence patient outcomes [6]. Systemic inflammation is central to plaque rupture and myocardial injury, and elevated inflammatory indices are consistently linked to heightened risk [7,8]. Concurrently, physiological reserve and nutritional status, often reflected by serum albumin levels, are crucial determinants of long-term survival [9]. Low serum albumin, a negative acute-phase reactant, is commonly observed in cardiovascular patients and is strongly associated with adverse outcomes, potentially reflecting chronic catabolism, hepatic dysfunction or compromised physiological resilience [10,11].
The platelet-to-albumin ratio (PAR) is a novel composite metric derived from two routinely measured blood components. Platelets are key cellular mediators of thrombosis and the inflammatory cascade in ACS [12,13], while albumin reflects overall systemic health and reserve [14,15]. As a composite index derived from platelet count and serum albumin, the PAR captures information on both the acute-phase inflammatory response and host nutritional standing. Examining these factors in combination may provide a more comprehensive reflection of the clinical state in ACS, as it reflects the balance between inflammatory triggers and systemic defences.
Recent investigations have begun to explore the role of the composite PAR metric in various ACS sub-cohorts and related complications. For instance, Hao et al. reported an association between PAR and major adverse cardiovascular events in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS) treated with percutaneous coronary intervention (PCI) [16]. Notably, their analysis did not reveal a significant association between PAR and all-cause mortality within that specific cohort. Other studies have linked PAR to complication risks: Ding et al. found an association between PAR and ventricular aneurysm formation in STEMI patients [17], and Wang et al. reported that PAR was associated with new-onset atrial fibrillation in elderly patients with acute STEMI [18]. Despite these findings suggesting a link between PAR and various intermediate outcomes, the comprehensive association between PAR and the critical endpoint of one-year all-cause mortality across the broader, heterogeneous ACS population remains incompletely elucidated and warrants focused investigation.
Therefore, given the composite biological relevance of PAR in integrating inflammatory and nutritional status, this retrospective cohort study hypothesizes that the PAR is significantly associated with one-year all-cause mortality in patients diagnosed with ACS.
Patients and methods
Study design and population
This single-centre, retrospective cohort study was conducted at the Affiliated Hospital of Guangdong Medical University, encompassing patients consecutively hospitalized with a diagnosis of ACS between January 2022 and December 2023. The diagnosis of ACS including STEMI, NSTEMI and UA was established by attending physicians based on clinical and biochemical evidence, in accordance with the 2021 Guidelines for the evaluation and diagnosis of chest pain [19]. This study received ethical approval from the Ethics Committee of the Affiliated Hospital of Guangdong Medical University (No. KT2025-283-01) and adhered to the principles of the Declaration of Helsinki. All participants provided written informed consent and the study was registered with the Chinese Clinical Trial Registry (ChiCTR2500113985). We excluded patients based on the following criteria: (1) lack of baseline platelet count or serum albumin data necessary for calculating the PAR or (2) absence of follow-up data regarding the primary outcome of one-year all-cause mortality.
Data collection
The following data were recorded in this study: clinical data (age, sex, body mass index (BMI), smoking, drinking), vital signs at admission (heart rate, systolic blood pressure and diastolic blood pressure), medical history (previous ACS, hypertension, dyslipidaemia, diabetes, previous stroke, atrial fibrillation), ACS type, laboratory data (white blood cells, red blood cells, haemoglobin, lymphocytes, monocytes, neutrophils, platelet, total cholesterol, triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, estimated glomerular filtration rate, haemoglobin A1c, albumin, creatinine, uric acid, potassium, sodium), echocardiography parameters (left ventricular ejection fraction), PCI and medication at discharge (nitrates/nicorandil, aspirin/indobufen, P2Y12 inhibitors, dual antiplatelet therapy (DAPT), statins, ezetimibe, calcium channel blocker, renin–angiotensin–aldosterone system inhibitors and beta-blockers). These data, spanning from January 2022 to December 2023, were retrospectively analysed in March 2026.
Definitions of PAR and outcomes
PAR was calculated as the platelet count (in ×109/L) divided by the serum albumin concentration (in g/L) [20,21]. The platelet count and serum albumin concentration were obtained from the first fasting venous blood test report after admission. The primary endpoint was one-year all-cause mortality. The secondary endpoints was a composite outcome, major adverse cardiovascular and cerebrovascular events (MACCEs), which included all-cause mortality, non-fatal myocardial infarction, ischemic stroke, rehospitalization for heart failure, revascularization or in-stent thrombosis of the target vessel, and malignant arrhythmia. Detailed definitions of all outcomes are provided in Supplementary Table S1.
Follow-up
Patients were followed for up to one year following hospital admission. Data on MACCE were ascertained exclusively from the hospital’s electronic medical records, and telephone follow-ups were solely employed to confirm the occurrence of primary outcome. Survival time was calculated from the date of admission until either the date of death or the censoring date at 365 days for surviving patients.
Statistical analysis
The distribution of continuous data was scrutinized through the Shapiro–Wilk test. Normal and non-normal variables were characterized as mean ± standard deviation (SD) or medians with interquartile ranges (IQRs), with comparisons performed using Student’s t-test or the Mann–Whitney U-test, as appropriate. For categorical variables, counts and percentages were reported. Comparisons across the PAR groups involved one-way analysis of variance (ANOVA) for continuous variables and the Chi-square test for categorical data.
The cumulative incidences of primary and secondary endpoints were estimated using the Kaplan–Meier method and compared via the log-rank test. To evaluate the association between PAR and the outcomes, we employed multivariable Cox proportional hazards regression models, treating PAR as both a continuous variable and a categorical variable (quartiles). Covariates were selected based on a priori clinical relevance and established evidence in ACS research. To minimize the risk of overfitting given the number of observed events, we developed a parsimonious model restricted to 10 key variables. These included established cardiovascular risk factors (age, sex, BMI, smoking, ACS type and hypertension) and evidence-based treatments (PCI, DAPT, beta-blockers and statins), all of which were forced into the models regardless of their univariable p values. We constructed three sequential models to assess these associations: (1) a crude model; (2) model 1, adjusted for demographics and lifestyle factors (age, sex, BMI and smoking); and (3) model 2, which further adjusted for ACS type, comorbidities (hypertension) and clinical treatments (PCI, DAPT, beta-blockers and statins). This consistent set of covariates was applied to both mortality and MACCE outcomes.
Restricted cubic spline (RCS) analysis with four knots (at the 5th, 35th, 65th and 95th percentiles) was performed to explore potential non-linear associations between PAR and the study outcomes, with knot placement optimized by minimizing the Akaike information criterion (AIC). These models were adjusted for the aforementioned confounders. When a non-linear relationship was identified, a two-piecewise linear regression model was employed to determine the threshold effect (inflection point). The 95% confidence intervals (CIs) for the inflection point were estimated using the bootstrap percentile method with 1000 resamples. In instances where bootstrap-based CIs were unstable or degenerated, the profile likelihood method was alternatively employed to ensure robust estimation.
Missing data were handled based on the mechanism and frequency of missingness. For BMI (56.6% missing), we primarily categorized it into three levels (<25 kg/m2, ≥25 kg/m2 and ‘missing’) to minimize potential bias. To address concerns regarding this handling, we performed two sensitivity analyses: first, by excluding BMI from the multivariable models entirely, and second, by using multiple imputation for the missing BMI values. For other covariates with missingness <9% (as shown in Supplementary Table S2), multiple imputation was consistently utilized.
Finally, a series of sensitivity analyses were conducted to evaluate the robustness of our findings. This included: (1) subgroup analyses across various clinical characteristics for all-cause mortality, visualized using forest plots; (2) RCS models stratified by ACS subtypes (NSTE-ACS undergoing PCI, STEMI, NSTEMI and UA) to assess the consistency of the association; and (3) formal interaction analyses (PAR × ACS type) via likelihood ratio tests to determine whether the ACS subtype significantly modified the observed associations.
All statistical analyses were conducted using R software version 4.4.2 (https://cran.r-project.org/). A two-sided p value < 0.05 was considered statistically significant.
Results
Study population characteristics
Following the application of the inclusion and exclusion criteria, a total of 1326 eligible ACS patients were included in the final analysis (Figure 1). The cohort had a median age of 67 years (IQR: 58–76), and 324 (24.4%) participants were female. Over the one-year follow-up period, the incidences of all-cause mortality and MACCEs were 10.3% and 21.1%, respectively. Details on the incidence and distribution of MACCE are provided in Supplementary Table S3. All participants were stratified into four groups based on PAR quartiles: Q1 (PAR ≤ 5.12, n = 335), Q2 (5.12 < PAR ≤ 6.08, n = 330), Q3 (6.08 < PAR ≤ 7.33, n = 331) and Q4 (PAR > 7.33, n = 330). Compared with patients in Q2 and Q3, those in Q1 and Q4 were significantly older (p < 0.001). Furthermore, participants in the Q1 and Q4 groups received DAPT and statins less frequently than those in the middle quartiles (Table 1). A comprehensive overview of all baseline characteristics, including vital signs and detailed laboratory parameters, is provided in Supplementary Table S4.
Figure 1.
![Flowchart detailing patient selection for ACS study: 1401 initial patients, 75 excluded, resulting in 1326 analyzed, divided into four quartiles. The flowchart illustrates the patient selection process for a study on Acute Coronary Syndrome (ACS) at the Affiliated Hospital of Guangdong Medical University (January 2022 - December 2023). It starts with 1401 patients (n=1401), showing exclusions: 2 for missing platelet count, 53 for serum albumin, and 25 for one-year mortality, leading to 1326 analyzed patients. These are further categorized into quartiles: Q1 (n=335, range [0.19, 5.12]), Q2 (n=330, range (5.12, 6.08]), Q3 (n=331, range (6.08, 7.33]), Q4 (n=330, range (7.33, 27.75]). Each quartile connects to the final cohort count.](https://www.ncbi.nlm.nih.gov/core/lw/2.0/html/tileshop_pmc/tileshop_pmc_inline.html?title=Click%20on%20image%20to%20zoom&p=PMC3&id=13403452_IANN_A_2704986_F0001_C.jpg)
Flowchart of the study. Overlapping data deficiencies were observed within the excluded cohort: one patient had missing values for platelet count, serum albumin and one-year mortality status; one patient had missing values for both platelet count and serum albumin; and two patients lacked both serum albumin and one-year mortality status. ACS: acute coronary syndrome.
Table 1.
Baseline characteristics according to quartiles of the PAR levels.
| Characteristic | Total (n = 1326) | Quartiles of PAR | p Value | |||
| Quartile 1 (n = 335) PAR ≤ 5.12 |
Quartile 2 (n = 330) 5.12 < PAR ≤ 6.08 |
Quartile 3 (n = 331) 6.08 < PAR ≤ 7.33 |
Quartile 4 (n = 330) PAR > 7.33 |
|||
| Clinical data | ||||||
| Age (years) | 67.0 (58.0, 76.0) | 70.0 (60.0, 78.0) | 64.0 (55.0, 74.0) | 65.0 (56.0, 74.0) | 68.0 (58.0, 79.0) | <0.001 |
| Female sex, n (%) | 324 (24.4) | 60 (17.9) | 74 (22.4) | 78 (23.6) | 112 (33.9) | <0.001 |
| BMI, n (%) | 0.036 | |||||
| Missing | 751 (56.6) | 189 (56.4) | 184 (55.8) | 173 (52.3) | 205 (62.1) | |
| <25 (kg/m2) | 391 (29.5) | 105 (31.3) | 103 (31.2) | 95 (28.7) | 88 (26.7) | |
| ≥25 (kg/m2) | 184 (13.9) | 41 (12.2) | 43 (13.0) | 63 (19.0) | 37 (11.2) | |
| Smoking, n (%) | 323 (24.4) | 82 (24.5) | 80 (24.2) | 91 (27.5) | 70 (21.2) | 0.300 |
| Drinking, n (%) | 117 (8.8) | 32 (9.6) | 27 (8.2) | 35 (10.6) | 23 (7.0) | 0.400 |
| Medical history, n (%) | ||||||
| Hypertension | 604 (45.6) | 144 (43.0) | 149 (45.2) | 159 (48.0) | 152 (46.1) | 0.600 |
| Dyslipidemia | 306 (23.1) | 67 (20.0) | 78 (23.6) | 87 (26.3) | 74 (22.4) | 0.300 |
| Diabetes | 384 (29.0) | 83 (24.8) | 97 (29.4) | 102 (30.8) | 102 (30.9) | 0.300 |
| Previous stroke | 28 (2.1) | 7 (2.1) | 7 (2.1) | 7 (2.1) | 7 (2.1) | >0.9 |
| ACS type, n (%) | 0.030 | |||||
| STEMI | 624 (47.1) | 159 (47.5) | 143 (43.3) | 158 (47.7) | 164 (49.7) | |
| NSTEMI | 609 (45.9) | 141 (42.1) | 161 (48.8) | 155 (46.8) | 152 (46.1) | |
| UA | 93 (7.0) | 35 (10.4) | 26 (7.9) | 18 (5.4) | 14 (4.2) | |
| Laboratory data | ||||||
| Platelet (×109/L) | 237 (198, 278) | 178 (154, 193) | 221 (204, 237) | 257 (238, 272) | 313 (283, 348) | <0.001 |
| Albumin (g/L) | 38.7 (35.7, 41.1) | 39.3 (37.1, 41.9) | 39.7 (36.9, 41.7) | 38.7 (36.0, 41.1) | 36.4 (32.2, 39.2) | <0.001 |
| PCI, n (%) | 922 (69.5) | 226 (67.5) | 237 (71.8) | 238 (71.9) | 221 (67.0) | 0.300 |
| Medication at discharge, n (%) | ||||||
| Aspirin/indobufen | 907 (68.4) | 225 (67.2) | 248 (75.2) | 226 (68.3) | 208 (63.0) | 0.009 |
| P2Y12 inhibitors | 1,117 (84.2) | 275 (82.1) | 288 (87.3) | 280 (84.6) | 274 (83.0) | 0.300 |
| DAPT | 0.020 | |||||
| Clopidogrel-based | 218 (16.4%) | 52 (15.5%) | 56 (17.0%) | 55 (16.6%) | 55 (16.7%) | |
| Ticagrelor-based | 654 (49.3%) | 156 (46.6%) | 186 (56.4%) | 166 (50.2%) | 146 (44.2%) | |
| No | 454 (34.2%) | 127 (37.9%) | 88 (26.7%) | 110 (33.2%) | 129 (39.1%) | |
| Statins | 1,094 (82.5) | 270 (80.6) | 284 (86.1) | 280 (84.6) | 260 (78.8) | 0.048 |
ACS: acute coronary syndrome; BMI: body mass index; DAPT: dual antiplatelet therapy; PCI: percutaneous coronary intervention; PAR: platelet to albumin ratio; STEMI: ST elevation myocardial infarction; UA: unstable angina; NSTEMI: non-ST elevation myocardial infarction.
Continuous variables are presented as mean ± SD or median (first quartile to third quartile). Categorical variables are presented as number (%).
Kaplan–Meier analysis for primary and secondary outcomes
One-year mortality rates across the PAR quartiles were 11.6% (Q1), 6.1% (Q2), 7.0% (Q3) and 16.7% (Q4), while MACCE rates followed a similar pattern (22.1%, 15.2%, 18.7% and 28.5%, respectively; Supplementary Figure S1). Kaplan–Meier curves revealed significant disparities in clinical outcomes across the quartiles (log-rank p < 0.001 for all; Figure 2). Specifically, patients in the second quartile (Q2) of PAR exhibited the lowest risk of death and MACCE. Compared with the Q2 group, the risk of these outcomes was significantly higher in both the lowest (Q1) and highest (Q4) quartiles.
Figure 2.
Kaplan–Meier survival curves in different PAR groups for (A) one-year all-cause mortality rates and (B) MACCE rates. MACCE: major adverse cardiovascular and cerebrovascular events; PAR: platelet-to-albumin ratio.
Associations between PAR and the outcomes
Cox regression analysis was conducted to evaluate the associations between PAR and one-year all-cause mortality, as well as MACCE (Table 2). After adjusting for potential confounding factors, each 1-unit increment in PAR (as a continuous variable) was associated with a 11% increased risk of one-year mortality (hazard ratio (HR) = 1.11, 95% CI: 1.05–1.17, p < 0.001) and a 7% increased risk of MACCE (HR = 1.07, 95% CI: 1.02–1.12, p = 0.002). In the categorical analysis using the second quartile (Q2) as the reference group, patients in the lowest (Q1) and highest (Q4) quartiles exhibited a heightened risk of adverse outcomes. Specifically, the risk of all-cause mortality was significantly elevated in Q4 (HR = 2.30, 95% CI: 1.37–3.84, p = 0.002), while a similar trend was observed in Q1, though it did not reach statistical significance (HR = 1.63, 95% CI: 0.95–2.81, p = 0.079). A consistent pattern was identified for MACCE, with significantly higher risk in Q4 (HR = 1.78, 95% CI: 1.26–2.51, p = 0.001) compared to the reference group.
Table 2.
Cox regression model.
| Events (rate, %) | Crude model | p Value | Adjusted model 1 | p Value | Adjusted model 2 | ||
|---|---|---|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | HR (95% CI) | p Value | ||||
| 1-Year mortality | |||||||
| PAR (continuous) | 137 (10.3) | 1.12 (1.06–1.18) | <0.001 | 1.11 (1.05–1.17) | <0.001 | 1.11 (1.05–1.17) | <0.001 |
| PAR | |||||||
| Q1 | 39 (11.6) | 2.00 (1.17–3.43) | 0.012 | 1.67 (0.97–2.87) | 0.064 | 1.63 (0.95–2.81) | 0.079 |
| Q2 | 20 (6.1) | 1 (ref.) | 1 (ref.) | 1 (ref.) | |||
| Q3 | 23 (7.0) | 1.15 (0.63–2.10) | 0.644 | 1.11 (0.61–2.02) | 0.743 | 1.08 (0.59–1.97) | 0.814 |
| Q4 | 55 (16.7) | 2.94 (1.76–4.91) | <0.001 | 2.50 (1.49–4.18) | <0.001 | 2.30 (1.37–3.84) | 0.002 |
| MACCE | |||||||
| PAR (continuous) | 280 (21.1) | 1.08 (1.03–1.13) | <0.001 | 1.07 (1.03–1.12) | <0.001 | 1.07 (1.02–1.12) | 0.002 |
| PAR | |||||||
| Q1 | 74 (22.1) | 1.57 (1.09–2.24) | 0.014 | 1.41 (0.98–2.03) | 0.061 | 1.38 (0.96–1.98) | 0.083 |
| Q2 | 50 (15.2) | 1 (ref.) | 1 (ref.) | 1 (ref.) | |||
| Q3 | 62 (18.7) | 1.28 (0.88–1.85) | 0.199 | 1.24 (0.86–1.81) | 0.253 | 1.18 (0.81–1.71) | 0.395 |
| Q4 | 94 (28.5) | 2.07 (1.47–2.91) | <0.001 | 1.90 (1.34–2.69) | <0.001 | 1.78 (1.26–2.51) | 0.001 |
CI: confidence interval; HR: hazard ratio; MACCE: major adverse cardiovascular and cerebrovascular events; PAR: platelet-to-albumin ratio.
Crude model: adjusted for none. Adjusted model 1: adjusted for age, sex, body mass index and smoking. Adjusted model 2: adjusted for age, sex, body mass index, smoking, acute coronary syndrome type, hypertension, percutaneous coronary intervention, dual antiplatelet therapy, beta-blockers and statins.
Nonlinear relationship between PAR and the outcomes
Within the Cox regression framework, RCS were employed to characterize the non-linear relationship between PAR and the study outcomes, after adjusting for potential confounders. As illustrated in Figure 3, a non-linear, U-shaped association was identified for both one-year all-cause mortality and MACCE (both p for non-linearity < 0.05). Further analysis using two-piecewise Cox regression identified specific inflection points for these associations (Table 3): 5.28 (95% CI: 4.86–9.86) for mortality and 5.29 (95% CI: 4.37–6.58) for MACCE. Below these thresholds, the PAR was inversely associated with the risk of adverse outcomes; specifically, for every 10-unit increase in the PAR, the adjusted HR were 0.06 (95% CI: 0.00–4.06) for one-year all-cause mortality and 0.13 (95% CI: 0.01–2.10) for MACCE. Conversely, above these inflection points, a positive correlation emerged, where each 10 units increase in PAR was linked to a markedly higher risk of mortality (HR = 4.47, 95% CI: 2.56–7.81) and MACCE (HR = 2.64, 95% CI: 1.68–4.17).
Figure 3.
Restricted cubic spline regression and two-piecewise linear regression for (A) one-year all-cause mortality and (B) MACCE. All models were adjusted for age, sex, body mass index, smoking, acute coronary syndrome type, hypertension, percutaneous coronary intervention, dual antiplatelet therapy, beta-blockers and statins. HR: hazard ratio; MACCE: major adverse cardiovascular and cerebrovascular events; PAR: platelet-to-albumin ratio.
Table 3.
Two-piecewise Cox proportional model.
| Adjusted HR per 10 units (95% CI) | p Value | |
|---|---|---|
| One-year mortality | ||
| Inflection point (K) | 5.28 (4.86–9.86) | |
| <K | 0.06 (0.00–4.06) | 0.193 |
| ≥K | 4.47 (2.56–7.81) | <0.001 |
| MACCE | ||
| Inflection point (K) | 5.29 (4.37–6.58) | |
| <K | 0.13 (0.01–2.10) | 0.149 |
| ≥K | 2.64 (1.68–4.17) | <0.001 |
CI: confidence interval; HR: hazard ratio; MACCE: major adverse cardiovascular and cerebrovascular events.
All models were adjusted for age, sex, body mass index, smoking, acute coronary syndrome type, hypertension, percutaneous coronary intervention, dual antiplatelet therapy, beta-blockers and statins.
Sensitivity analyses
The results remained robust when BMI was either excluded from the multivariable adjustments or when missing BMI values were handled via multiple imputation (Supplementary Tables S5 and S6). For all-cause mortality, subgroup analyses consistently showed that the association between PAR and mortality remained stable across various clinical characteristics, as illustrated in the forest plots (Supplementary Figures S2 and S3).
Furthermore, RCS models indicated that the non-linear association between PAR and clinical outcomes was generally consistent across STEMI and NSTEMI subgroups; however, the association did not reach statistical significance in the NSTEMI subgroups (p for overall > 0.05). In the specific subgroup of patients with NSTE-ACS undergoing PCI, PAR was not significantly associated with all-cause mortality or MACCE (all p for overall > 0.05). Detailed results for these analyses are provided in Supplementary Figures S4 and S5. Despite these variations, formal interaction analyses (PAR × ACS) yielded non-significant interaction p values for both all-cause mortality (p = 0.457) and MACCE (p = 0.963).
Discussion
To date, clinical investigations exploring the association between the PAR and outcomes in patients with ACS remain limited. In this study, we evaluated this relationship and found that PAR was independently associated with both one-year all-cause mortality and MACCE. A key finding of our analysis is the observed U-shaped relationship, where patients in both the lowest and highest PAR ranges exhibited a higher risk of adverse clinical events.
This non-linear association extends the findings of Hao et al. [16], who reported a link between PAR and MACCE in NSTE-ACS patients treated with PCI but did not observe a significant association with all-cause mortality. Regarding NSTE-ACS patients undergoing PCI, our results were consistent with theirs concerning all-cause mortality. However, regarding MACCE, the discrepancy between our findings and those of Hao et al. might be attributed to the limited sample size of this subgroup in our study, or variations in the definition of MACCE across studies.
To evaluate the consistency of these associations across different clinical presentations, we performed sensitivity analyses across different ACS subtypes (STEMI, NSTEMI and UA). Although the non-linear patterns were generally observed across all subtypes, the associations reached statistical significance primarily in STEMI patients and the overall cohort, likely due to larger sample sizes. Crucially, formal interaction analyses (PAR × ACS) yielded non-significant interaction p values for both outcomes, indicating that the ACS subtype did not significantly modify the observed associations.
The divergence in results regarding all-cause mortality between our study and Hao et al. may also stem from differences in study populations. While their study focused exclusively on NSTE-ACS patients undergoing PCI, our cohort encompasses a more diverse ACS population. Given that the acute phase of STEMI is often characterized by more intense systemic inflammatory responses, the inclusion of a broader spectrum of ACS subtypes likely captured a wider range of physiological status. Consequently, this heterogeneity underscores that the U-shaped association between PAR and clinical outcomes is a robust feature across the ACS spectrum, rather than being confined to a single clinical subtype.
The observed independent U-shaped association suggests that both extremes of the PAR spectrum are linked to distinct pathophysiological states in ACS. For individuals in the highest PAR group, the elevated risk is likely associated with a synergistic state of heightened platelet reactivity and systemic inflammation. Elevated platelet levels reflect a pro-thrombotic environment and increased plaque instability; activated platelets interact with the vascular endothelium, triggering the release of pro-inflammatory cytokines and the recruitment of leukocytes into the vascular wall, which accelerates vascular injury [13]. Concurrently, low albumin levels often reflect a heightened acute-phase inflammatory response. Under systemic inflammation, the liver prioritizes the synthesis of acute-phase proteins, such as C-reactive protein, while inhibiting albumin synthesis [22,23]. This reduction in albumin leads to a weakening of its anti-inflammatory and antioxidant effects, which further increases oxidative stress and promotes endothelial dysfunction. Consequently, the combination of excessive thrombotic activity and compromised vascular protection is strongly associated with the occurrence of adverse events.
Conversely, the increased risk associated with a low PAR manifests a different clinical profile, potentially related to a state of haematological imbalance or reduced physiological reserve. Evidence from large-scale ACS cohorts indicates that a decline in platelet counts or the presence of thrombocytopenia is significantly associated with an increased risk of mortality [24,25]. In this context, a low PAR, driven by a reduced platelet count, may reflect a compromised response to acute stress or an underlying systemic vulnerability. Furthermore, as demonstrated in existing studies on threshold effects, the protective association of albumin may reach a plateau or even exhibit a paradoxical relationship when its concentration deviates significantly from the biological equilibrium [9,10]. In such cases, a very low PAR relates to depleted platelet reserves or abnormally high albumin levels, reflecting a loss of physiological homeostasis. Under these conditions, the body’s compensatory mechanisms may become overwhelmed or imbalanced during the acute coronary event.
The clinical implications of our findings provide a more nuanced insight into the association between the PAR and one-year outcomes in patients with ACS. Rather than a simple linear relationship, we identified specific inflection points – 5.28 for one-year all-cause mortality and 5.29 for MACCE – which may delineate an optimal homeostatic range. Within this interval, the interplay between platelet-mediated thrombotic activity and albumin-dependent systemic protection appears most favourably balanced. Consequently, deviations from these thresholds facilitate the identification of distinct clinical profiles, where either extreme of the PAR spectrum is associated with a significantly higher risk of adverse events. These identified thresholds offer potential clinical utility for risk stratification. Specifically, these values could serve as practical cut-offs to categorize ACS patients into different risk profiles at admission, distinguishing between those within the ‘optimal homeostatic range’ and those beyond these thresholds who may warrant more vigilant clinical assessment or intensified therapeutic strategies. By providing a quantitative basis for identifying patients at elevated risk, these inflection points represent a promising, easily calculable tool that may enhance the granularity of clinical decision-making in the management of ACS.
Several limitations of this study should be acknowledged. First, the retrospective, single-centre design may limit the generalizability of our findings to more diverse populations. Second, although we adjusted potential confounders, residual confounding from unmeasured factors cannot be entirely excluded. For instance, we lacked data on the specific duration and longitudinal adjustments of DAPT regimens. As recently highlighted by others [26], these dynamic factors significantly influence the balance between ischemic and bleeding risks. Finally, while we identified significant associations and specific inflection points, the nature of this study precludes the establishment of direct causality. Further prospective, multicentre studies are required to validate these findings and to explore the underlying biological mechanisms in more detail.
Conclusions
PAR levels show a U-shaped association with one-year all-cause mortality and MACCE in patients with ACS. Our findings demonstrate that inflection points exist within the clinical range of PAR, which may represent an optimal homeostatic state. For clinicians, these points provide a practical reference for risk stratification at admission. Patients with PAR levels deviating from this range may require more vigilant clinical assessment and more intensive management. Future research should prioritize prospective, multicentre studies to validate these specific inflection points in diverse populations. Additionally, mechanistic studies are needed to clarify the biological pathways linking extreme PAR values to adverse clinical outcomes.
Supplementary Material
Acknowledgements
The authors are grateful to Siyao Wei, Shuyi Zhou and Ziqi Lin from The First School of Clinical Medicine, Guangdong Medical University, for their invaluable contributions to the data collection process.
Funding Statement
This study was supported by the Natural Science Foundation of Guangdong Province (Grant No. 2024A1515013119), the National Natural Science Foundation of China (Grant No. 82370281) and the National Key Research and Development Program Special Projects (Grant No. 2024YFA1802203).
Ethical approval
Ethical approval for this retrospective study was granted by the Ethics Committee of Affiliated Hospital of Guangdong Medical University (ethical approval number: KT2025-283-01); the research complies with the Declaration of Helsinki.
Consent form
All patients provided written informed consent to participate. The trial was registered at the Chinese Clinical Trial Registry (registration number: ChiCTR2500113985; registration date: 05 December 2025).
Disclosure statement
No potential conflict of interest was reported by the author(s).
Open scholarship
This article has earned the Center for Open Science badge for Preregistered. The materials are openly accessible at: https://www.chictr.org.cn/hvshowproject.html?id=289570&v=1.0.
Data availability statement
All research data can be obtained from the corresponding author upon reasonable request.
References
- 1.Bergmark BA, Mathenge N, Merlini PA, et al. Acute coronary syndromes. Lancet. 2022;399(10332):1347–1358. doi: 10.1016/s0140-6736(21)02391-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bhatt DL, Lopes RD, Harrington RA.. Diagnosis and treatment of acute coronary syndromes: a review. JAMA. 2022;327(7):662–675. doi: 10.1001/jama.2022.0358. [DOI] [PubMed] [Google Scholar]
- 3.Byrne RA, Rossello X, Coughlan JJ, et al. 2023 ESC Guidelines for the management of acute coronary syndromes: developed by the task force on the management of acute coronary syndromes of the European Society of Cardiology (ESC) (vol 13, pg 55, 2024). Eur Heart J Acute Cardiovasc Care. 2024;13(5):455–455. doi: 10.1093/ehjacc/zuad156. [DOI] [PubMed] [Google Scholar]
- 4.Sawano M, Kohsaka S, Ishii H, et al. One-year outcome after percutaneous coronary intervention for acute coronary syndrome – an Analysis of 20,042 patients from a Japanese Nationwide Registry. Circ J. 2021;85(10):1756–1767. doi: 10.1253/circj.CJ-21-0098. [DOI] [PubMed] [Google Scholar]
- 5.Batty J, Qiu W, Gu S, et al. One-year clinical outcomes in older patients with non-ST elevation acute coronary syndrome undergoing coronary angiography: an analysis of the ICON1 study. Int J Cardiol. 2019;274:45–51. doi: 10.1016/j.ijcard.2018.09.086. [DOI] [PubMed] [Google Scholar]
- 6.Ajoolabady A, Pratico D, Lin L, et al. Inflammation in atherosclerosis: pathophysiology and mechanisms. Cell Death Dis. 2024;15(11):817. doi: 10.1038/s41419-024-07166-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Li J, Wei W, Xiao L, et al. Association of the systemic inflammation response index with the one-year all-cause mortality in elderly patients with acute coronary syndrome: a retrospective cohort study. J Inflamm Res. 2026;19:593688. doi: 10.2147/JIR.S593688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Li Q, Ma X, Shao Q, et al. Prognostic impact of multiple lymphocyte-based inflammatory indices in acute coronary syndrome patients. Front Cardiovasc Med. 2022;9:811790. doi: 10.3389/fcvm.2022.811790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Li X, Zhang Y, He Y, et al. J-shaped association between serum albumin levels and long-term mortality of cardiovascular disease: experience in National Health and Nutrition Examination Survey (2011–2014). Front Cardiovasc Med. 2022;9:1073120. doi: 10.3389/fcvm.2022.1073120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhu L, Chen M, Lin X.. Serum albumin level for prediction of all-cause mortality in acute coronary syndrome patients: a meta-analysis. Biosci Rep. 2020;40(1):BSR20190881. doi: 10.1042/BSR20190881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Manolis AA, Manolis TA, Melita H, et al. Low serum albumin: a neglected predictor in patients with cardiovascular disease. Eur J Intern Med. 2022;102:24–39. doi: 10.1016/j.ejim.2022.05.004. [DOI] [PubMed] [Google Scholar]
- 12.Davì G, Patrono C.. Platelet activation and atherothrombosis. N Engl J Med. 2007;357(24):2482–2494. doi: 10.1056/NEJMra071014. [DOI] [PubMed] [Google Scholar]
- 13.Gawaz M, Langer H, May AE.. Platelets in inflammation and atherogenesis. J Clin Invest. 2005;115(12):3378–3384. doi: 10.1172/JCI27196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fanali G, di Masi A, Trezza V, et al. Human serum albumin: from bench to bedside. Mol Aspects Med. 2012;33(3):209–290. doi: 10.1016/j.mam.2011.12.002. [DOI] [PubMed] [Google Scholar]
- 15.Belinskaia DA, Jenkins RO, Goncharov NV.. Serum albumin in health and disease: from comparative biochemistry to translational medicine. Int J Mol Sci. 2023;24(18):13725. doi: 10.3390/ijms241813725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Hao P, Feng S, Suo M, et al. Platelet to albumin ratio: a risk factor related to prognosis in patients with non-ST-segment elevation acute coronary syndrome undergoing percutaneous coronary intervention. Int J Cardiol. 2024;395:131588. doi: 10.1016/j.ijcard.2023.131588. [DOI] [PubMed] [Google Scholar]
- 17.Ding H, Chai R, Yin Y, et al. Platelet-to-albumin ratio is an independent predictor for ventricular aneurysm formation in ST segment elevation myocardial infarction. Coron Artery Dis. 2025;36(2):117–125. doi: 10.1097/mca.0000000000001445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Wang F, Sun Y, Lu Y, et al. Predictive value of platelet-to-albumin ratio combined with the C2HEST score for new-onset atrial fibrillation in elderly patients with acute ST-segment elevation myocardial infarction. BMC Cardiovasc Disord. 2024;24(1):521. doi: 10.1186/s12872-024-04200-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Gulati M, Levy PD, Mukherjee D, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR guideline for the evaluation and diagnosis of chest pain: executive summary: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2021;144(22):e368–e454. doi: 10.1161/CIR.0000000000001030. [DOI] [PubMed] [Google Scholar]
- 20.Xiao L, Li F, Sheng Y, et al. Predictive value analysis of albumin-related inflammatory markers for short-term outcomes in patients with in-hospital cardiac arrest. Expert Rev Clin Immunol. 2025;21(2):249–257. doi: 10.1080/1744666x.2024.2399700. [DOI] [PubMed] [Google Scholar]
- 21.Qiu J, Yu Y, Wang Z, et al. Comprehensive analysis of the prognostic value of pre-treatment nutritional indicators in elderly rectal cancer patients. Sci Rep. 2024;14(1):22078. doi: 10.1038/s41598-024-73123-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Moshage HJ, Janssen JA, Franssen JH, et al. Study of the molecular mechanism of decreased liver synthesis of albumin in inflammation. J Clin Invest. 1987;79(6):1635–1641. doi: 10.1172/JCI113000. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Sheinenzon A, Shehadeh M, Michelis R, et al. Serum albumin levels and inflammation. Int J Biol Macromol. 2021;184:857–862. doi: 10.1016/j.ijbiomac.2021.06.140. [DOI] [PubMed] [Google Scholar]
- 24.Zito A, Landi A, Milzi A, et al. In-hospital platelet count dynamics in patients with acute coronary syndrome. Rev Esp Cardiol. 2026;79(3):257–269. doi: 10.1016/j.rec.2025.09.002. [DOI] [PubMed] [Google Scholar]
- 25.Wang TY, Ou FS, Roe MT, et al. Incidence and prognostic significance of thrombocytopenia developed during acute coronary syndrome in contemporary clinical practice. Circulation. 2009;119(18):2454–2462. doi: 10.1161/CIRCULATIONAHA.108.827162. [DOI] [PubMed] [Google Scholar]
- 26.De Filippo O, Piroli F, Bruno F, et al. De-escalation of dual antiplatelet therapy for patients with acute coronary syndrome after percutaneous coronary intervention: a systematic review and network meta-analysis. BMJ Evid Based Med. 2024;29(3):171–186. doi: 10.1136/bmjebm-2023-112476. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All research data can be obtained from the corresponding author upon reasonable request.