Association between platelet-to-lymphocyte ratio and 28-day all-cause mortality in patients with asymptomatic coronary artery disease: a cohort study Authors: Jianfeng Zhao, Zhitao Li, Shi Fu, Mingjie Fu, Qinghui Fu, Shuiqiao Fu Published: 2025 ABSTRACT AbstractBackgroundRecent evidence suggests a link between platelet-to-lymphocyte ratio (PLR) and coronary artery disease (CAD) outcomes. However, studies on asymptomatic CAD (ACAD) and short-term outcomes like 28-day mortality are limited. This research will explore the correlation between PLR and 28-day mortality in ACAD patients and assess a potential dose–response relationship.MethodsThis retrospective cohort study leverages data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. We included patients diagnosed with ACAD and collected baseline characteristics, laboratory test results, and clinical data during hospitalization. The primary endpoint was 28-day all-cause mortality. Analytical approaches included multivariable Cox proportional hazards regression models, stratified analysis, restricted cubic splines (RCS), and threshold effect analyses to evaluate the association between PLR and 28-day all-cause mortality. Given the skewed distribution of PLR values, natural logarithm (ln) transformation was applied for statistical analyses. ResultsOur study encompassed 4,563 ACAD patients, of whom 71.7% were male, with a mean age of 71.8 years. Overall, 12.8% of patients experienced 28-day mortality. Analysis revealed a J-shaped curve relationship between PLR and 28-day all-cause mortality. Threshold analysis indicated that participants with lnPLR < 4.308 had a Hazard Ratio (HR) of 0.814 for 28-day all-cause mortality (95% CI 0.627–1.057; P = 0.122), whereas those with lnPLR > 4.308 had an HR of 1.441 (95% CI 1.245–1.668; P < 0.001). Below the threshold, increasing lnPLR did not significantly correlate with changes in 28-day mortality risk. Conversely, above this threshold, each unit increase in lnPLR was associated with a 44.1% increased risk of 28-day mortality.ConclusionAmong adult patients with ACAD in the United States, there exists a J-shaped relationship between PLR and 28-day all-cause mortality. Elevated PLR levels are associated with adverse outcomes in ACAD patients.Supplementary InformationThe online version contains supplementary material available at 10.1186/s40001-025-03494-3.Keywords: Platelet-to-lymphocyte ratio (PLR),; Asymptomatic coronary heart disease (ACAD); 28-day all-cause mortality; Cohort study Introduction Coronary artery disease (CAD) is one of the leading causes of death worldwide, with its incidence increasing year by year, drawing significant public attention [1]. Asymptomatic coronary artery disease (ACAD) is a type of CAD where patients do not exhibit typical angina symptoms, but have significant stenosis or obstruction in the coronary arteries. This disease is clinically insidious and challenging to diagnose, yet it has a substantial impact on the prognosis of cardiovascular events. The high rates of occurrence and mortality associated with this condition are linked to various risk factors, including hypertension, high cholesterol, tobacco use, diabetes, physical inactivity, and obesity [2]. To tackle the challenges posed by asymptomatic CHD, it is crucial to develop effective public health policies. The risk of developing CHD can be significantly reduced by adopting healthier lifestyles, keeping blood pressure and cholesterol levels within recommended ranges, quitting smoking, and engaging in regular physical activity [3]. Furthermore, the early detection and management of individuals at high risk are essential strategies for mitigating the overall impact of the disease. Research has shown that traditional markers of inflammation, such as C-reactive protein (CRP) and white blood cell count (WBC), have limited effectiveness in predicting short-term mortality [4]. For instance, in patients experiencing cardiogenic shock, CRP levels were found to correlate with the 30-day all-cause mortality rate from the third to the eighth day, but not significantly on the day of admission [5]. Additionally, in the context of coronary heart disease, inflammatory markers like CRP and IL-6 are closely linked to components of metabolic syndrome and severe ventricular arrhythmias, highlighting the role of inflammation in increasing cardiovascular disease risk [6]. However, the predictive value of these markers alone remains limited, and they should be used alongside other biomarkers or clinical characteristics to enhance predictive accuracy and improve clinical application. The platelet-to-lymphocyte ratio (PLR) is an emerging composite biomarker that has shown significant promise in evaluating inflammatory responses across a range of diseases [7]. By quantifying the activation levels of platelets and reflecting the exhaustion status of lymphocytes, PLR provides a deeper understanding of chronic inflammatory responses. In chronic inflammatory diseases, variations in PLR can indicate the severity of systemic inflammation and help predict the onset of infections and other related health issues [8]. Additionally, platelets play a vital role in inflammation and immune regulation [9]; they are not only involved in hemostasis and wound healing but also in recognizing pathogenic microorganisms and secreting immune-regulatory cytokines, which support various immune functions [10]. This dual role makes PLR an important marker for assessing the risk of cardiovascular diseases, particularly in patients with chronic inflammatory conditions, where platelet activation is closely associated with ongoing endothelial cell activation [11]. In clinical practice, PLR is a simple, stable, and easily accessible inflammatory marker that has been utilized to predict the prognosis of various diseases [7, 12]. Accumulating evidence from big cohort studies has demonstrated that PLR is highly predictive of adverse clinical outcomes in patients with acute coronary syndrome [12–14]. Nevertheless, its predictive power in populations with asymptomatic CHD has not been completely clarified or validated. This research, drawing upon the MIMIC-IV database, has created a large-scale retrospective cohort and has systematically investigated the dose–response relationship between PLR and the 28-day all-cause mortality rate in asymptomatic CHD patients. The objective is to furnish solid, evidence-based medical evidence to refine risk stratification for this specific demographic. Methods Study design and population The investigation utilized data from the Medical Information Mart for Intensive Care IV (MIMIC-IV version 3.0) database, which includes records of 364,627 patients hospitalized at the Beth Israel Deaconess Medical Center in the United States between 2008 and 202215 [15]. We obtained the data for the ICD-10 code I2510, which corresponds to “atherosclerotic heart disease of the native coronary artery without angina pectoris”, from the patient diagnosis records in the database. This cohort was designated as the primary research group for the study of asymptomatic coronary heart disease, comprising a total of 41,550 cases. After excluding 33,946 individuals who were not first-time admissions to the Intensive Care Unit (ICU), 1196 patients whose ICU stays were shorter than 24 h, and 1845 patients with incomplete data on platelet and lymphocyte counts, the final cohort consisted of 4563 patients. These participants were divided into four groups based on the quartiles of the platelet-to-lymphocyte ratio (PLR); group Q2 included 1140 patients, while the other three groups each contained 1141 patients. The selection process for study participants is illustrated in Fig. 1, highlighting the thoroughness of the sampling method and the reliability of the analytical results, thus providing a solid foundation for exploring the relationship between PLR and the risk of mortality over a 28-day period. The study was conducted using the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [16]. Flowchart of patient selection To ensure patient confidentiality, all identifiers that could reveal personal information were removed. The Institutional Review Board at Beth Israel Deaconess Medical Center concluded that explicit consent from each individual was unnecessary due to the comprehensive verification of patient records within the MIMIC-IV database. This study was carried out in compliance with the ethical standards set forth in the Declaration of Helsinki. Data collection and PLR definition To gain access to this database, we completed the "CITI Data or Specimens Only Research" training course available on the National Institutes of Health website, and the authors were granted permission to retrieve data from the database for research purposes (certificate number: 64490942). We extracted various variables from the MIMIC-IV database, encompassing demographic information, comorbidities, laboratory outcomes, disease severity metrics, vital sign measurements, and discharge statuses. Following univariate analysis and considering clinical relevance, we chose age, gender, race, myocardial infarction, congestive heart failure, chronic pulmonary disease, dementia, kidney disease, SAPS II score, OASIS score, APACHE III score, Charlson Comorbidity Index, respiratory rate, diastolic blood pressure, blood oxygen saturation, hemoglobin, anion gap, blood urea nitrogen, creatinine, chloride concentration, lactate, blood glucose, international normalized ratio, and base excess as covariates for further multivariate analysis [17]. Asymptomatic coronary artery disease was characterized as primary coronary atherosclerotic heart disease in the absence of angina pectoris [18], and laboratory test parameters included lymphocyte count and platelet count. We utilized the initial laboratory test results within 24 h of ICU admission for statistical analysis. PLR was defined as the platelet-to-lymphocyte ratio [19]. Patients were categorized into quartiles based on their PLR values, with Q1 (0.063–57.303), Q2 (57.317–93.069), Q3 (93.077–177.551), and Q4 (177.647–3422.222). The study's endpoint was the all-cause mortality rate observed during the 28-day hospitalization period. Statistical analysis We undertook a comprehensive analysis of the distribution characteristics of the platelet-to-lymphocyte ratio (PLR). To assess the normality of continuous variables, we employed the Shapiro–Wilk test. Due to the observed right-skewed nature of PLR levels, we implemented a natural logarithm transformation (ln) to facilitate the attainment of a normal distribution [20]. For variables conforming to a normal distribution, we reported the data as mean ± standard deviation, which were subsequently compared using one-way analysis of variance. In instances of skewed distributions within continuous data, the median and interquartile range were utilized, and the Kruskal–Wallis H test was conducted to evaluate the results. Categorical or dichotomous variables were expressed as absolute counts alongside percentages, with the Chi-square test applied for analysis. Moreover, cases with incomplete data regarding platelet and lymphocyte counts were promptly excluded. Remaining missing values were addressed through multivariable chained imputation employing fully conditional specifications; particularly, the results post-imputation were found to be consistent with the originally reported outcomes [21]. We established three distinct models to investigate the relationship between PLR and 28-day all-cause mortality. Model 1 was unadjusted for any covariates, whereas Model 2 incorporated adjustments for various factors, including age, gender, race, prior myocardial infarction (MI), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), dementia (DEM), renal disease (RD), Acute Physiology Score II (APS II), Acute Physiology Score III (APS III), Acute Illness Severity Score (OASIS), and Charlson Comorbidity Index (CCI). Model 3 further refined Model 2 by incorporating adjustments for additional variables such as respiratory rate (RR), diastolic blood pressure (DBP), oxygen saturation (SPO2), hemoglobin (HB), anion gap (AG), blood urea nitrogen (BUN), blood chloride, lactate, glucose, international normalized ratio (INR), and base excess. To evaluate the association between varying levels of PLR and the risk of 28-day all-cause mortality among patients, we utilized Kaplan–Meier survival curves. An inflection point analysis was conducted to determine the optimal PLR threshold associated with the lowest risk of 28-day mortality. To bolster the robustness of our findings, stratified analyses were performed, taking into account variables such as age, gender, race, lactate levels, and the presence of conditions like myocardial infarction, congestive heart failure, COPD, and renal disease. All statistical analyses were executed using R statistical software (version 4.2.2), available at http://www.R-project.org (R Foundation), alongside a complimentary statistical analysis platform (version 2.1, Beijing, China, (http://www.clinicalscientists.cn/freestatistics)). A two-sided p-value below 0.05 was deemed statistically significant. Results Baseline characteristics description This study included 4,563 asymptomatic coronary heart disease patients, of which 71.7% were male, with Caucasians being the majority. The average age of the overall population was 71.8 years old, the median ICU stay was 2.3 days, and the 28-day all-cause mortality rate was 12.8%. As shown in Table 1, with the increase of PLR levels, there was a significant trend of increasing age in patients, and the Charlson Comorbidity Index, blood pressure, body temperature, carbon dioxide partial pressure, platelet count, hematocrit, creatinine, blood urea nitrogen, etc., showed an upward trend. Conversely, oxygen saturation, white blood cells, lymphocytes, serum chloride, lactate levels, etc., showed a downward trend, indicating that PLR levels were closely related to the patient's basic inflammatory load and the complexity of comorbidities. Table 1Baseline characteristics of patients with asymptomatic coronary heart disease according to PLR quartileVariablesTotal (n = 4563)PLRQ1 (n = 1141)Q2 (n = 1140)Q3 (n = 1141)Q4 (n = 1141)P valueDemographic Age, ± SD71.8 ± 11.270.2 ± 10.670.4 ± 11.072.1 ± 11.674.7 ± 11.2 < 0.001 Gender, n (%) < 0.001 Male3272 (71.7)854 (74.8)859 (75.4)809 (70.9)750 (65.7) Female1291 (28.3)287 (25.2)281 (24.6)332 (29.1)391 (34.3) Race, n (%)0.981 White3008 (65.9)757 (66.3)747 (65.5)751 (65.8)753 (66) Non-White1555 (34.1)384 (33.7)393 (34.5)390 (34.2)388 (34)Comorbidities MI, n (%) < 0.001 No2438 (53.4)654 (57.3)654 (57.4)565 (49.5)565 (49.5) Yes2125 (46.6)487 (42.7)486 (42.6)576 (50.5)576 (50.5) CHF, n (%) < 0.001 No2574 (56.4)757 (66.3)771 (67.6)584 (51.2)462 (40.5) Yes1989 (43.6)384 (33.7)369 (32.4)557 (48.8)679 (59.5) PVD, n (%)0.007 No3819 (83.7)948 (83.1)984 (86.3)962 (84.3)925 (81.1) Yes744 (16.3)193 (16.9)156 (13.7)179 (15.7)216 (18.9) DEM, n (%) < 0.001 No4305 (94.3)1117 (97.9)1097 (96.2)1067 (93.5)1024 (89.7) Yes258 (5.7)24 (2.1)43 (3.8)74 (6.5)117 (10.3) COPD, n (%) < 0.001 No3520 (77.1)933 (81.8)936 (82.1)876 (76.8)775 (67.9) Yes1043 (22.9)208 (18.2)204 (17.9)265 (23.2)366 (32.1) DMwCC, n (%) < 0.001 No3524 (77.2)944 (82.7)911 (79.9)855 (74.9)814 (71.3) Yes1039 (22.8)197 (17.3)229 (20.1)286 (25.1)327 (28.7) RD, n (%) < 0.001 No3272 (71.7)902 (79.1)888 (77.9)782 (68.5)700 (61.3) Yes1291 (28.3)239 (20.9)252 (22.1)359 (31.5)441 (38.7)Severity scores CCI6.7 ± 2.75.9 ± 2.35.9 ± 2.47.0 ± 2.77.9 ± 2.7 < 0.001 APS III,46.6 ± 22.843.3 ± 22.242.2 ± 21.546.5 ± 22.454.3 ± 23.1 < 0.001 APS II39.7 ± 12.640.3 ± 12.438.0 ± 12.138.7 ± 12.841.8 ± 12.5 < 0.001 OASIS33.1 ± 8.633.0 ± 8.332.0 ± 8.332.7 ± 8.734.7 ± 9.1 < 0.001Vital signs HR, beats/min81.3 ± 13.581.3 ± 11.080.0 ± 11.180.2 ± 14.683.6 ± 16.2 < 0.001 SBP, mmHg114.7 ± 13.8111.6 ± 9.9113.4 ± 11.7116.8 ± 15.1117.0 ± 16.7 < 0.001 DBP, mmHg60.8 ± 10.058.1 ± 7.959.6 ± 9.062.1 ± 11.063.3 ± 11.0 < 0.001 MBP, mmHg77.0 ± 9.374.7 ± 6.976.0 ± 8.278.4 ± 10.378.9 ± 10.8 < 0.001 RR, beats/min19.0 ± 3.318.5 ± 3.018.4 ± 2.819.0 ± 3.220.0 ± 3.7 < 0.001 Temperature, °C36.8 ± 0.436.7 ± 0.436.7 ± 0.436.8 ± 0.436.8 ± 0.4 < 0.001 Spo2, %97.0 ± 2.097.5 ± 1.897.3 ± 1.896.8 ± 1.996.4 ± 2.1 < 0.001Laboratory parameters WBC, × 10⁹/L16.1 ± 9.619.1 ± 14.215.9 ± 6.514.9 ± 7.114.3 ± 7.7 < 0.001 Lymphocytes, × 10⁹/L1.5 (0.9, 2.2)2.7 (2.1, 3.5)1.8 (1.4, 2.2)1.3 (1.0, 1.6)0.7 (0.5, 0.9) < 0.001 Monocytes, × 10⁹/L0.6 (0.4, 0.9)0.6 (0.3, 0.9)0.5 (0.3, 0.9)0.7 (0.4, 1.0)0.7 (0.4, 1.0) < 0.001 Neutrophils, × 10⁹/L9.3 (6.6, 13.1)9.8 (7.1, 13.2)9.0 (6.2, 12.3)8.9 (6.2, 12.8)9.7 (6.7, 13.9) < 0.001 PLT, × 10⁹/L160.1 ± 78.5113.4 ± 45.5136.4 ± 47.3171.2 ± 60.6219.6 ± 101.5 < 0.001 HCT, %29.4 ± 6.527.4 ± 5.528.8 ± 5.930.6 ± 7.230.7 ± 6.7 < 0.001 HB, mg/dL9.6 ± 2.29.0 ± 1.99.5 ± 2.010.0 ± 2.49.9 ± 2.3 < 0.001 Glucose, mmol/L132.5 (120.0, 153.4)132.0 (123.1, 145.5)130.5 (120.6, 145.0)133.0 (118.6, 158.5)138.0 (114.0, 177.7) < 0.001 BUN, mg/dL21.0 (15.0, 34.0)17.0 (13.0, 25.0)18.0 (14.0, 26.0)22.0 (16.0, 36.0)31.0 (19.0, 49.0) < 0.001 Cr, mg/dL1.1 (0.9, 1.6)1.0 (0.8, 1.3)1.0 (0.8, 1.4)1.2 (0.9, 1.8)1.4 (0.9, 2.3) < 0.001 INR1.6 ± 0.91.6 ± 0.91.5 ± 0.61.5 ± 0.81.7 ± 1.1 < 0.001 PT, s17.2 ± 9.818.0 ± 11.016.4 ± 6.316.3 ± 8.318.0 ± 12.4 < 0.001 APTT, s33.3 (29.0, 47.0)34.1 (29.8, 45.7)32.7 (28.9, 41.4)32.9 (28.7, 49.6)33.4 (28.6, 55.6)0.002 PH7.3 ± 0.17.3 ± 0.17.3 ± 0.17.3 ± 0.17.3 ± 0.1 < 0.001 PO2, mmHg96.8 ± 43.1100.9 ± 36.3103.2 ± 39.196.5 ± 45.786.5 ± 48.5 < 0.001 PCO2, mmHg36.1 ± 7.234.4 ± 5.635.6 ± 5.736.6 ± 7.637.7 ± 9.0 < 0.001 Calcium, mmol/L8.6 ± 0.78.4 ± 0.78.5 ± 0.78.7 ± 0.78.7 ± 0.8 < 0.001 Chloride, mmol/L105.8 ± 5.5107.8 ± 4.1106.9 ± 4.8105.2 ± 5.3103.1 ± 6.4 < 0.001 Sodium, mmol/L139.0 ± 4.4138.9 ± 3.5138.7 ± 4.0139.4 ± 4.3139.1 ± 5.40.002 Potassium, mmol/L4.8 ± 0.84.7 ± 0.74.7 ± 0.74.8 ± 0.84.9 ± 1.0 < 0.001 Lactate, mmol/L2.4 (1.8, 3.3)2.9 (2.1, 4.1)2.4 (1.9, 3.2)2.2 (1.6, 3.0)2.1 (1.4, 3.1) < 0.001 AG, mmol/L15.2 ± 5.113.9 ± 4.813.9 ± 4.515.7 ± 5.117.2 ± 5.1 < 0.001 BE, mmol/L− 3.0 (− 5.0, − 1.0)− 4.0 (− 6.0, − 2.0)− 3.0 (− 5.0, − 1.0)− 2.0 (− 5.0, 0.0)− 2.0 (− 6.0, 0.0) < 0.001Outcome Death within ICU, n (%) < 0.001 No4249 (93.1)1088 (95.4)1105 (96.9)1063 (93.2)993 (87) Yes314 (6.9)53 (4.6)35 (3.1)78 (6.8)148 (13) Death within hospital, n (%) < 0.001 No4104 (89.9)1074 (94.1)1085 (95.2)1027 (90)918 (80.5) Yes459 (10.1)67 (5.9)55 (4.8)114 (10)223 (19.5) ICU Los, days2.3 (1.4, 4.2)2.2 (1.3, 4.0)2.1 (1.3, 3.3)2.4 (1.5, 4.5)3.0 (1.8, 5.6) < 0.001 Hospital Los, days7.8 (5.3, 12.7)7.7 (5.6, 11.7)6.9 (5.1, 10.5)7.7 (5.1, 12.6)9.1 (5.6, 15.8) < 0.001 Death within 28 days, n (%) < 0.001 No3979 (87.2)1068 (93.6)1071 (93.9)998 (87.5)842 (73.8) Yes584 (12.8)73 (6.4)69 (6.1)143 (12.5)299 (26.2)MI myocardial infarction, CHF congestive heart failure, PVD peripheral vascular disease, DEM dementia, COPD chronic obstructive pulmonary disease, DMwCC diabetes mellitus with chronic complications, RD renal disease, CCI Charlson Comorbidity Index, APS III Acute Physiology Score III, APS II Acute Physiology Score II, OASIS Outcome and Severity of Illness Score, HR heart rate, SBP systolic blood pressure, DBP diastolic blood pressure, MBP mean blood pressure, RR respiratory rate, Spo2 oxygen saturation, WBC white blood cell count, PLT platelets, HCT hematocrit, HB hemoglobin, BUN blood urea nitrogen, Cr creatinine, INR international normalized ratio, PT prothrombin time, APTT activated partial thromboplastin time, PO2 partial pressure of oxygen, PCO2 partial pressure of carbon dioxide, AG anion gap, BE base excess, PLR platelet/lymphocyte ratio Baseline characteristics of patients with asymptomatic coronary heart disease according to PLR quartile MI myocardial infarction, CHF congestive heart failure, PVD peripheral vascular disease, DEM dementia, COPD chronic obstructive pulmonary disease, DMwCC diabetes mellitus with chronic complications, RD renal disease, CCI Charlson Comorbidity Index, APS III Acute Physiology Score III, APS II Acute Physiology Score II, OASIS Outcome and Severity of Illness Score, HR heart rate, SBP systolic blood pressure, DBP diastolic blood pressure, MBP mean blood pressure, RR respiratory rate, Spo2 oxygen saturation, WBC white blood cell count, PLT platelets, HCT hematocrit, HB hemoglobin, BUN blood urea nitrogen, Cr creatinine, INR international normalized ratio, PT prothrombin time, APTT activated partial thromboplastin time, PO2 partial pressure of oxygen, PCO2 partial pressure of carbon dioxide, AG anion gap, BE base excess, PLR platelet/lymphocyte ratio Univariate regression analysis The univariate regression analysis indicated that patient age, gender, race, heart rate, respiratory rate, oxygen saturation, blood glucose, hematocrit, platelet count, albumin, blood urea nitrogen, creatinine, calcium, potassium, lymphocyte count, international normalized ratio, prothrombin time, activated partial thromboplastin time, lactate, pH value, acute physiology score, and Charlson Comorbidity Index were all significantly correlated with the 28-day all-cause mortality rate (all P values < 0.05) (Supplementary Table 1). These findings provide a basis for further studies on the risk of death in patients with asymptomatic coronary heart disease. Dose–effect relationship between PLR and mortality risk Multivariate Cox regression models revealed the independent predictive value of PLR for prognosis. In the multivariable model adjusted for demographic characteristics, comorbidity factors, laboratory parameters, and vital signs, for every 1-unit increase in the natural logarithm of PLR (lnPLR), the mortality risk increased by 38.4% (HR = 1.384, 95% CI 1.254–1.527) (Table 2). The quartile analysis with Q2 as the reference group showed that the fully adjusted risks in groups Q3 and Q4 increased by 62.5 and 176.5%, respectively (P < 0.001) (Table 3). Table 2Multivariate Cox regression analysis was used to assess the association between PLR and 28-day mortalityVariableModel 1Model 2Model3HR (95%CI)P valueHR (95%CI)P valueHR (95%CI)P valuelnPLR Continuous variable1.938 (1.778 ~ 2.113) < 0.0011.481 (1.352 ~ 1.623) < 0.0011.384 (1.254 ~ 1.527) < 0.001PLR (Quartile) Q21(Ref)1(Ref)1(Ref) Q11.062 (0.765 ~ 1.476)0.71831.068 (0.764 ~ 1.492)0.70011.065 (0.766 ~ 1.482)0.7073 Q32.146 (1.61 ~ 2.86) < 0.0011.528 (1.139 ~ 2.049)0.00471.625 (1.215 ~ 2.172)0.001 Q44.811 (3.703 ~ 6.251) < 0.0012.378 (1.804 ~ 3.135) < 0.0012.765 (2.114 ~ 3.615) < 0.001Model 1: unadjustedModel 2: age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCIModel 3: Model 2 + RR, DBP, SPO2, HB, AG, BUN, Chloride, lactate, glucose, INR, BEHR hazard risk, CI confidence interval Multivariate Cox regression analysis was used to assess the association between PLR and 28-day mortality Model 1: unadjusted Model 2: age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCI Model 3: Model 2 + RR, DBP, SPO2, HB, AG, BUN, Chloride, lactate, glucose, INR, BE HR hazard risk, CI confidence interval Table 3Threshold effect analysis of relationships between lnPLR and 28-day all-cause mortalityItemHR (95%CI)P valueThreshold4.308 (3.781,4.836) < 4.3080.814 (0.627,1.057)0.122 ≥ 4.3081.441 (1.245,1.668) < 0.001Likelihood ratio test–0.001We adjusted for age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCI, RR, DBP, SPO2, HB, AG, BUN, chloride, lactate, glucose, INR, BE. HR hazard risk, CI confidence interval Threshold effect analysis of relationships between lnPLR and 28-day all-cause mortality We adjusted for age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCI, RR, DBP, SPO2, HB, AG, BUN, chloride, lactate, glucose, INR, BE. HR hazard risk, CI confidence interval Nonlinear association and survival analysis The restricted cubic spline model revealed a significant J-shaped association between lnPLR and mortality rate (nonlinear P < 0.001), with the minimum risk threshold at lnPLR = 4.308. When lnPLR exceeded this threshold, the risk of death increased exponentially (Fig. 2). Kaplan–Meier survival curves further confirmed the dose–effect relationship (P < 0.0001), with the 28-day cumulative survival rate in group Q4 being significantly lower than that in the other three groups (Fig. 3). J-shaped relationships between lnPLR and 28-day all-cause mortality in patients with asymptomatic coronary heart disease. Adjusted for age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCI, RR, DBP, SPO2, HB, AG, BUN, chloride, lactate, glucose, INR, BE. Only 99% of the data is shown Kaplan–Meier survival analysis curves for 28-day all-cause mortality Subgroup analysis The forest plot analysis (Fig. 4) indicated that PLR was a risk factor for 28-day all-cause mortality in asymptomatic coronary artery disease. The results of the subgroup analysis suggested that this outcome was consistent across various subgroups. There was significant population heterogeneity in the prognostic value: in elderly patients and patients with cardiorenal comorbidities, a stronger association was observed. Notably, no significant association was found in the subgroup with lactate ≥ 5 mmol/L (Supplementary Table 2). Forest plots for different subgroup analyses of HR for the association between lnPLR and 28-day all-cause mortality in patients with asymptomatic coronary heart disease. The effect size was adjusted for age, gender, race, MI, CHF, COPD, DEM, RD, APS II, APS III, OASIS, CCI, RR, DBP, SPO2, HB, AG, BUN, chloride, lactate, glucose, INR, BE. HR Hazard Risk, CI confidence interval Discussion This large-scale cohort study, which included 4,563 asymptomatic coronary artery disease (ACAD) patients, provides evidence linking the platelet-to-lymphocyte ratio (PLR) to 28-day all-cause mortality. We discovered a novel nonlinear J-shaped relationship between PLR and prognosis in these patients. Through restricted cubic spline analysis and threshold analysis, we identified a turning point at lnPLR = 4.308, beyond which the risk of mortality increases. These findings enhance our understanding of the prognostic significance of PLR in ACAD patients. Our results are consistent with previous studies on the predictive value of PLR; the threshold-based J-shaped relationship aligns with Li et al., who found that a PLR greater than 151.28 predicted major adverse cardiovascular events (MACE) in patients with AMI [22], which corresponds with the platelet activation thresholds necessary for stable thrombus formation [23]. The relationship between PLR and 28-day mortality was particularly pronounced in subgroups of elderly patients (≥ 65 years), and those with cardiorenal comorbidities, while no significant association was found in the subgroup with lactate levels ≥ 5 mmol/L (P = 0.682). The increased risk associated with cardiorenal comorbidities aligns with the findings of Zeng et al., who reported PLR might be a potential biomarker to predict 5-year all-cause mortality in CKD patients11, potentially due to uremia-induced platelet hyperreactivity [24]. Interestingly, our J-shaped curve results differ from previous studies, where PLR was linearly associated with poor prognosis in patients with symptomatic coronary artery disease [25]. This suggests that there may be different pathophysiological mechanisms between the asymptomatic and symptomatic stages of the disease. ACAD is characterized by chronic endothelial dysfunction and intermittent low-grade inflammation, while symptomatic CAD is marked by IL-6-driven acute plaque instability [6]. Previous studies that employed arbitrary PLR categorization, such as tertiles, may have overlooked the nonlinear dynamics that can be better captured through spline models [26]. However, the lack of association observed in the hyperlactatemia subgroup (≥ 5 mmol/L) was contrary to our initial assumptions. This is an interesting phenomenon, but no similar content has been reported in other literature. This discrepancy may be due to mitochondrial dysfunction caused by high lactate levels masking the thrombotic effects of PLR [27]. Additionally, the small sample size of this subgroup may limit the statistical power of the study results. Platelet-to-lymphocyte ratio (PLR), as a composite biomarker reflecting inflammation and thrombotic status, has shown good predictive value across a variety of disease fields. However, the optimal cutoff defining an elevated PLR differs among clinical conditions, and a universally accepted threshold has yet to be established. Research indicates that a PLR exceeding 252.4 is linked to an increased mortality risk in patients with heart failure [28]. In a study involving 288 patients who experienced acute myocardial infarction, a higher PLR (exceeding 178) is associated with negative outcomes such as cardiac rupture, acute heart failure, and elevated all-cause mortality [12]. Our study indicates that ACAD patients with a lnPLR ≥ 4.308 show an increasing trend in short-term mortality. Considering that ACAD is an early stage of acute coronary syndrome (ACS), we recommend combining the PLR with other inflammatory markers (such as neutrophil-to-lymphocyte ratio (NLR), C-reactive protein (CRP)) and clinical indicators (such as APACHE II score, SOFA score) to promptly intervene in high-risk populations and prevent the progression of asymptomatic coronary heart disease to symptomatic coronary heart disease. The PLR is closely related to the risk of thrombosis. Studies have shown that an increased PLR is associated with poor outcomes in various cardiovascular diseases, including acute myocardial infarction and chronic occlusive lesions [29, 30]. An elevated PLR may reflect an enhanced state of inflammation and thrombosis in the body, a mechanism that could be related to increased platelet activation, increased release of inflammatory factors, and interactions between platelets and leukocytes [31, 32]. Furthermore, as a comprehensive indicator of inflammation and thrombosis, an increased PLR suggests that a patient is in a prothrombotic or inflammatory state, thereby increasing the risk of thrombosis. Antiplatelet therapy is a key strategy for preventing and treating thrombosis. However, bleeding is an important complication of antiplatelet treatment, especially in patients with acute coronary syndrome (ACS), where the risk of bleeding is closely related to mortality rates. Research by Spadafora et al. indicates that in-hospital bleeding significantly affects post-discharge antithrombotic strategies and long-term outcomes [33]. Eikelboom et al. also found that the 30-day mortality rate in patients with severe bleeding was five times higher than in those without bleeding, and the more severe the bleeding, the higher the risk of death [34]. In clinical practice, it is necessary to find the optimal balance between the risks of bleeding and thrombosis. By individualized assessment, adjusting drug dosages, controlling underlying diseases, and dynamic monitoring, the safety and effectiveness of antiplatelet therapy can be effectively improved. In the future, with the conduct of more high-quality studies, the role of PLR in the management of cardiovascular diseases will be further clarified. Strengths and limitations The present study presents several strengths. Firstly, it employs restricted cubic splines to objectively identify PLR thresholds in ACAD, effectively addressing biases that can arise from arbitrary categorization [35]. Secondly, it incorporates sequential adjustments for 23 demographic, clinical, and laboratory covariates, which help to minimize confounding factors. Thirdly, the focus on asymptomatic CAD is significant, as it fills a critical evidence gap in the realm of primary prevention. Lastly, the study has important clinical implications: the low cost and accessibility of PLR enhance its utility in resource-limited settings; integrating PLR with GRACE scores could improve risk stratification for intensive lipid-lowering therapies [36, 37]; and monitoring PLR levels over time may assist in guiding the de-escalation of P2Y12 inhibitors in elderly patients [38]. However, the study also has several limitations. Firstly, the single-center design and the predominantly Caucasian cohort may limit the generalizability of the findings. Secondly, the reliance on a single PLR measurement taken within 24 h of admission restricts the ability to assess dynamic trends. Thirdly, like any retrospective study, unknown and unmeasured factors cannot be controlled. For instance, the study population may include patients who are comatose or unable to express symptoms, which raises questions about how the diagnosis of ACAD can be accurately defined and how data quality can be controlled. These factors could potentially impact the results, and the diagnosis of asymptomatic coronary artery disease is challenging, and the lack of validation through coronary CT or angiography in this study represents a limitation. Therefore, large-scale, multicenter, prospective cohort studies are urgently needed to address these limitations and to validate the prognostic significance of PLR in populations with asymptomatic coronary artery disease. Conclusion PLR can serve as a prognostic indicator for the 28-day mortality risk in patients with ACAD. The J-shaped association between PLR and the 28-day mortality risk may assist clinicians in stratifying patient management and serves as a reference. Future validation of this finding will require larger-scale prospective cohort studies. Supplementary Information Supplementary material 1 Acknowledgements The authors express gratitude to the Clinical Scientists' team for their invaluable contributions and extend appreciation to the Free Statistics Analysis Platform for their expertise in statistical analysis. Moreover,thank Kimi and DeepSeek for the help in polishing the thesis.Their involvement has significantly elevated the quality of this study. Author contributions Jianfeng Zhao and Zhitao Li conceived this study. Shuiqiao Fu designed the study. Qinghui Fu acquired and analyzed the data. Mingjie Fu and Shi Fu contributed analysis tools. Jianfeng Zhao wrote the paper. Zhitao Li were of immense help in the preparation of the manuscript. All authors read and approved the final manuscript. Funding No funding. Data availability The copies of the datasets used in this study are available from the MIMIC database. Declarations Ethics approval and consent to participate This study adhered to the principles outlined in the Declaration of Helsinki. Furthermore, it is important to note that the MIMIC-IV database received approval from the Massachusetts Institute of Technology and the Beth Israel Deaconess Medical Center. The copies of the datasets used in this study are available from the MIMIC database. Consequently, this research trial did not require ethical approval or the acquisition of informed consent. The Clinical Research Ethics Committee of the First Affiliated Hospital of Zhejiang University School determined that the study did not require ethical approval and consent for this study. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Qinghui Fu, Email: 2513107@zju.edu.cn. Shuiqiao Fu, Email: 2200048@zju.edu.cn. References Associated Data Supplementary Materials Supplementary material 1 Data Availability Statement The copies of the datasets used in this study are available from the MIMIC database.