Abstract
Objective
This study aimed to delineate the relationship between serum uric acid (UA) and 28-day and 90-day mortality among patients with acute myocardial infarction complicated by cardiogenic shock (AMI-CS) receiving veno-arterial extracorporeal membrane oxygenation (VA-ECMO).
Methods
In this retrospective, single-center cohort study, we consecutively enrolled 142 patients supported with VA-ECMO for AMI-CS. Baseline UA was treated as the exposure of interest, and 28-day and 90-day mortality were prespecified primary outcomes. Associations between UA and mortality were evaluated using Cox proportional-hazards models, Kaplan–Meier estimates, receiver operating characteristic curve (ROC), and restricted cubic spline (RCS) regression.
Results
Among 142 patients, the median baseline UA level was 452.00 μmol/L (361.00–561.75 μmol/L), and 28-day and 90-day mortality were 33.8% and 40.8%, respectively. After adjustment for potential confounders, each 1 μmol/L increment in UA remained associated with higher hazards of 28-day mortality (HR: 1.002, 95% CI: 1.000–1.003; P = 0.038) and 90-day mortality (HR: 1.002, 95% CI: 1.001–1.004; P = 0.005). Per one standard-deviation increase in UA, the hazards of 28-day and 90-day mortality increased by 37.9% (HR: 1.379, 95% CI: 1.019–1.867; P = 0.038) and 49.4% (HR: 1.494, 95% CI: 1.126–1.982; P = 0.005), respectively. Relative to the low-UA group, the high-UA group exhibited substantially higher risks of both 28-day (HR: 2.499, 95% CI: 1.335–4.677; P = 0.004) and 90-day mortality (HR: 2.262, 95% CI: 1.222–4.188; P = 0.009). ROC analyses indicated a moderate discriminatory performance of UA for 28-day and 90-day mortality, with AUCs of 0.631 (95% CI: 0.532–0.729; P = 0.011) and 0.621 (95% CI: 0.527–0.714; P = 0.015), respectively. Kaplan–Meier curves demonstrated significant separation in cumulative mortality between UA strata, with worse survival in the high-UA group (log-rank P = 0.001). RCS analyses further supported a significant linear association between UA and both endpoints (P for nonlinearity >0.05).
Conclusions
Elevated UA was associated with increased 28-day and 90-day mortality after VA-ECMO in AMI-CS. Given the single-center, retrospective, and observational design, these findings should be interpreted cautiously and require external validation before UA can be considered for clinical risk assessment.
Keywords: acute myocardial infarction, cardiogenic shock, extracorporeal membrane oxygenation, mortality, uric acid
1. Introduction
Cardiogenic shock (CS) remains a severe syndrome in acute cardiovascular medicine. Its pathophysiological hallmark is profound pump failure culminating in sustained tissue hypoperfusion and impaired oxygen delivery, thereby precipitating multiorgan dysfunction and death (1). Despite advances in emergent revascularization strategies, critical care techniques and mechanical circulatory support, overall outcomes remain poor, with in-hospital and short-term mortality persistently hovering around 40%–50% (2). Among CS etiologies, acute myocardial infarction complicated by cardiogenic shock (AMI-CS) is the most prevalent and is characterized by rapid clinical deterioration and marked heterogeneity, representing a continuing challenge in contemporary cardiovascular critical care. For refractory CS unresponsive to conventional therapy, veno-arterial extracorporeal membrane oxygenation (VA-ECMO) has emerged as a temporising form of mechanical support capable of rapidly restoring systemic perfusion and buying time for myocardial recovery or downstream bridging strategies (3, 4). However, several randomized trials and subsequent meta-analyses have not demonstrated a definitive survival benefit of VA-ECMO in AMI-CS, implying that although ECMO can acutely stabilize hemodynamics, longer-term outcomes may remain constrained by risk factors that are insufficiently recognized or inadequately targeted (5).
Beyond hemodynamic collapse, the CS cascade is tightly intertwined with metabolic disequilibrium, inflammatory activation and endothelial dysfunction, all of which may materially shape disease trajectories and prognosis (6). Uric acid (UA), the terminal product of purine metabolism, is widely viewed as an integrative readout of redox balance and metabolic homeostasis. Under physiological conditions, UA exerts antioxidant effects; yet in pathological states, hyperuricemia may promote cardiovascular injury via convergent mechanisms, including amplification of oxidative stress, propagation of inflammatory signalling, impairment of endothelial function and activation of the renin–angiotensin–aldosterone system (RAAS) (6, 7). Epidemiological and clinical cohort studies have consistently linked elevated UA to hypertension, coronary artery disease, heart failure (HF) and cardiovascular mortality (8–10). A large meta-analysis reported that each 1 mg/dL increase in serum UA is associated with approximately 20% higher risk of adverse cardiovascular events and 9% higher risk of all-cause mortality (10). In addition, a retrospective study of 23,475 patients suggested that UA may function as a metabolic marker for fatal AMI, with a risk threshold below the conventional diagnostic cut-off for hyperuricemia (optimal cut-off 5.70 mg/dL; fatal AMI incidence 2.6% below vs. 4.8% above; P < 0.0001) (11). Yet existing work has largely focused on general coronary populations or chronic HF cohorts, leaving the metabolic perturbations of AMI-CS—a severe and clinically complex state—relatively underexplored. This gap is particularly salient among AMI-CS patients supported with VA-ECMO, in whom extracorporeal circulation, ischemia–reperfusion injury, renal dysfunction and hyperinflammation may reshape UA kinetics, rendering its prognostic implications more complex. Under ECMO, conventional hemodynamic indices may be partially corrected, whereas metabolic and inflammatory signatures may provide additional prognostic information.
To date, systematic evidence delineating the relationship between UA and post-ECMO mortality risk in AMI-CS—and its dose–response form—remains limited. Against this backdrop, we conducted a single-center retrospective cohort study to interrogate the association between baseline UA and 28-day and 90-day mortality following VA-ECMO in AMI-CS, with the aim of exploring whether baseline UA is associated with short- and intermediate-term mortality in this clinically severe cohort.
2. Methods
2.1. Study population
This was a single-center retrospective cohort study. We consecutively enrolled adult patients (≥18 years) who received VA-ECMO for AMI-CS at Linyi People's Hospital between January 2021 and December 2025. Inclusion criteria were: (1) age ≥18 years; and (2) fulfilment of diagnostic criteria for AMI-CS. Exclusion criteria were: (1) no ECMO support; (2) missing baseline serum UA data; (3) patients with documented high-purine load before baseline UA assessment; or (4) loss to follow-up or refusal of follow-up. Ultimately, 142 patients were included in the final analysis. The study complied with the Declaration of Helsinki, was approved by the Ethics Committee of Linyi People's Hospital (Approval No. 202603-H-003), and since this was a retrospective study, the informed consent of all patients was waived by the Ethics Committee of Linyi People's Hospital.
2.2. Data collection and definitions
Clinical data were extracted from the electronic medical record system, including demographics, anthropometrics, medical history and comorbidities, severity indices, coronary angiography and intervention variables, laboratory measures, and ECMO-related treatment characteristics. Demographic variables included age, sex and smoking history; smoking was defined as any prior smoking irrespective of current smoking status (12). Anthropometric and vital-sign variables comprised body mass index (BMI = weight [kg]/height [m2]), systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR). Medical history and comorbidities included hypertension, diabetes, and prior myocardial infarction. Hypertension was defined as a prior diagnosis or SBP/DBP ≥ 140/90 mmHg during hospitalization (13); diabetes was defined as a prior diagnosis or fasting plasma glucose ≥7.0 mmol/L or glycated hemoglobin A1c (HbA1c) ≥6.5% during hospitalization (14). Clinical severity was characterized using the Society for Cardiovascular Angiography and Interventions (SCAI) shock staging, with inclusion restricted to stage C or higher (15). Angiography/intervention variables included number of diseased vessels, number of culprit vessels, number of stents, TIMI flow grade (0–3), and time to reperfusion. Laboratory variables included left ventricular ejection fraction (LVEF), lactate, white blood cell count (WBC), hemoglobin, platelet count (PLT), albumin, creatinine (Cr), baseline serum UA, potassium, sodium, calcium, glucose, total cholesterol, triglycerides, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), fibrinogen, high-sensitivity cardiac troponin T (hs-cTnT), creatine kinase-MB (CK-MB), and N-terminal pro-B-type natriuretic peptide (NT-proBNP). ECMO-related variables included cardiopulmonary resuscitation (CPR) prior to ECMO initiation, timing and location of ECMO cannulation, concomitant intra-aortic balloon pump (IABP) support, and durations of ECMO support, mechanical ventilation, IABP, and continuous renal replacement therapy (CRRT).
2.3. Measurement and stratification of UA
Serum UA was measured using a standard biochemical analyzer via the uricase–peroxidase method and reported in μmol/L. Baseline UA was defined as the available UA value measured within 24 h before VA-ECMO initiation. UA strata were assigned before subsequent ICU feeding strategies were considered, and nutritional support during VA-ECMO followed institutional critical-care protocols without routine high-purine or purine-containing supplementation. UA was analyzed as a continuous variable and also standardized as: standardized UA = (UA − mean)/standard deviation (SD). Furthermore, UA was categorized using receiver operating characteristic (ROC)-derived cut-offs based on the maximal Youden index for predicting 28-day and 90-day mortality. The corresponding apparent cut-offs were 530 μmol/L and 561 μmol/L, respectively. These cut-offs were regarded as exploratory data-driven thresholds rather than externally validated clinical decision limits. Accordingly, for descriptive and secondary categorical analyses, patients were stratified into low- and high-UA groups using the 28-day cut-off of 530 μmol/L and the 90-day cut-off of 561 μmol/L.
2.4. Follow-up and ascertainment of mortality
Follow-up information was obtained through review of electronic medical records (including rehospitalizations and outpatient/emergency visits) and structured telephone follow-up, capturing time of death and occurrence of death events. Survival time was defined as the interval from VA-ECMO initiation to the occurrence of an endpoint event or the end of follow-up. For descriptive analyses, patients were grouped by 28-day mortality (yes, n = 48; no, n = 94) and by 90-day mortality (yes, n = 58; no, n = 84).
2.5. Statistical analysis
Analyses were performed using SPSS version 26.0 and R version 4.4.3. Continuous variables were assessed for normality using the Shapiro–Wilk test. Normally distributed variables were presented as mean ± SD and compared using independent-samples t tests; non-normally distributed variables were presented as median (interquartile range) and compared using the Mann–Whitney U test. Categorical variables were summarized as counts (percentages) and compared using the χ2 test or Fisher's exact test, as appropriate.
Univariable Cox regression analysis was used to evaluate the correlation between each variable and the 28-day and 90-day mortality. Then, the variables with a P value <0.05 were selected for inclusion in the multivariate Cox regression analysis and three multivariate regression analysis models were constructed (while considering the collinearity among the variables): for 28-day mortality, Model 1: unadjusted; Model 2: adjusted for TIMI 3 vs. 0–2 only; Model 3: adjusted for TIMI 3 vs. 0–2, LVEF, lactate, creatinine, and Nt-ProBNP; for 90-day mortality, Model 1: unadjusted; Model 2: adjusted for TIMI 3 vs. 0–2 only; Model 3: adjusted for TIMI 3 vs. 0–2, BMI, LVEF, lactate, creatinine, and Nt-ProBNP. Furthermore, based on the full-adjustment model, a multivariate Cox regression analysis was conducted to explore the multivariate stratified association between UA and the 28-day and 90-day mortality risks in the IABP subgroup (yes or no). Results are reported as hazard ratios (HRs) with 95% confidence intervals (CIs). ROC curves were used to evaluate the discriminatory performance of UA for 28-day and 90-day mortality, quantified by the area under the curve (AUC). Because the ROC-derived cut-offs were generated and initially evaluated within the same cohort, bootstrap internal validation with 1,000 resamples was performed to assess the stability of the exploratory cut-offs and discrimination metrics. In each bootstrap sample, the optimal UA cut-off was re-estimated using the maximal Youden index, and the AUC, sensitivity, and specificity were calculated. Bootstrap distributions were summarized as medians with 95% CI. Kaplan–Meier survival curves were constructed and compared between UA strata using the log-rank test. Restricted cubic spline (RCS) regression, with covariate adjustment, was applied to assess potential non-linear associations between UA and mortality risk; nonlinearity was assessed by comparing the model with spline terms to the model with only the linear term using a likelihood ratio test. All tests were two-sided, and P < 0.05 was considered statistically significant.
3. Results
3.1. Clinical characteristics stratified by serum UA
As summarized in Table 1, patients were dichotomized into a low-UA group (≤530 μmol/L) and a high-UA group (>530 μmol/L) according to the exploratory ROC-derived cut-off for 28-day mortality. Baseline UA in the overall cohort was 452.00 μmol/L (361.00–561.75 μmol/L). Under the 530 μmol/L stratification, median UA was 381.00 μmol/L (314.25–456.75 μmol/L) in the low-UA group and 638.50 μmol/L (577.75–719.00 μmol/L) in the high-UA group. Compared with the low-UA group, patients with elevated UA were younger, exhibited a lower proportion of TIMI grade 2, and had higher creatinine and triglyceride concentrations (all P < 0.05). Both 28-day and 90-day mortality were higher in the high-UA group than in the low-UA group (P = 0.002 and 0.009, respectively). No other baseline variables differed significantly between groups (P > 0.05).
Table 1.
Clinical characteristics grouped by UA.
| Variables | Total population | Low UA (≤530) | High UA (>530) | P value | Low UA (≤561) | High UA (>561) | P value |
|---|---|---|---|---|---|---|---|
| N | 142 | 98 | 44 | 106 | 36 | ||
| Age, years | 57.54 ± 11.53 | 59.06 ± 11.66 | 54.14 ± 10.57 | 0.018 | 59.09 ± 11.49 | 52.94 ± 10.50 | 0.005 |
| Sex, n (%) | 0.056 | 0.088 | |||||
| Male | 112 (78.9) | 73 (74.5) | 39 (88.6) | 80 (75.5) | 32 (88.9) | ||
| Female | 30 (21.1) | 25 (25.5) | 5 (11.4) | 26 (24.5) | 4 (11.1) | ||
| Smoking, n (%) | 66 (46.5) | 46 (46.9) | 20 (45.5) | 0.870 | 49 (46.2) | 17 (47.2) | 0.918 |
| Hypertension, n (%) | 53 (37.3) | 37 (37.8) | 16 (36.4) | 0.874 | 38 (35.8) | 15 (41.7) | 0.533 |
| Diabetes, n (%) | 45 (31.7) | 32 (32.7) | 13 (29.5) | 0.713 | 33 (31.1) | 12 (33.3) | 0.806 |
| Previous MI, n (%) | 18 (12.7) | 14 (14.3) | 4 (9.1) | 0.390 | 14 (13.2) | 4 (11.1) | 1.000 |
| SCAI shock state, n (%) | 0.807 | 0.693 | |||||
| C | 16 (11.3) | 10 (10.2) | 6 (13.6) | 11 (10.4) | 5 (13.9) | ||
| D | 80 (56.3) | 55 (56.1) | 25 (56.8) | 59 (55.7) | 21 (58.3) | ||
| E | 46 (32.4) | 33 (33.7) | 13 (29.5) | 36 (34.0) | 10 (27.8) | ||
| Resuscitation before ECMO, n (%) | 53 (37.3) | 35 (35.7) | 18 (40.9) | 0.554 | 42 (39.6) | 11 (30.6) | 0.331 |
| BMI, kg/m2 | 24.80 (22.67, 27.08) | 24.63 (22.39, 26.45) | 25.53 (22.86, 27.68) | 0.107 | 24.49 (22.26, 26.73) | 25.92 (24.01, 27.67) | 0.031 |
| SBP, mmHg | 70.00 (0.00, 85.25) | 70.00 (40.00, 86.00) | 69.50 (0.00, 82.00) | 0.403 | 70.00 (0.00, 86.00) | 74.50 (15.00, 85.00) | 0.616 |
| DBP, mmHg | 45.00 (20.00, 60.00) | 45.00 (20.00, 60.00) | 45.00 (0.00, 52.75) | 0.153 | 42.00 (0.00, 59.25) | 48.00 (7.50, 55.50) | 0.964 |
| HR, beats/min | 76.00 (0.00, 109.25) | 76.00 (19.50, 109.25) | 74.00 (0.00, 109.00) | 0.492 | 68.50 (0.00, 108.25) | 83.50 (0.00, 110.00) | 0.554 |
| Diseased vessels ≥2, n (%) | 84 (59.2) | 57 (58.2) | 27 (61.4) | 0.720 | 61 (57.5) | 23 (63.9) | 0.504 |
| Culprit vessels ≥2, n (%) | 52 (36.6) | 34 (34.7) | 18 (40.9) | 0.477 | 36 (34.0) | 16 (44.4) | 0.259 |
| Stents implanted ≥2, n (%) | 28 (19.7) | 17 (17.3) | 11 (25.0) | 0.289 | 18 (17.0) | 10 (27.8) | 0.160 |
| TIMI grade, n (%) | 0.048 | 0.241 | |||||
| 0 | 8 (5.6) | 3 (3.1) | 5 (11.4) | 5 (4.7) | 3 (8.3) | ||
| 1 | 1 (0.7) | 0 (0.0) | 1 (2.3) | 0 (0.0) | 1 (2.8) | ||
| 2 | 13 (9.2) | 11 (11.2) | 2 (4.5) | 11 (10.4) | 2 (5.6) | ||
| 3 | 120 (84.5) | 84 (85.7) | 36 (81.8) | 90 (84.9) | 30 (83.3) | ||
| Time to revascularization, h | 7.00 (0.00, 24.00) | 6.50 (3.50, 16.00) | 8.50 (5.00, 48.00) | 0.138 | 6.00 (3.88, 16.75) | 9.00 (5.00, 48.00) | 0.080 |
| LVEF, % | 33.92 ± 9.63 | 34.78 ± 10.24 | 32.00 ± 7.91 | 0.113 | 35.01 ± 10.14 | 30.69 ± 7.16 | 0.007 |
| Lactate, mmol/L | 5.20 (4.00, 24.00) | 5.00 (3.30, 7.50) | 5.95 (3.43, 10.18) | 0.118 | 5.15 (3.30, 7.75) | 5.65 (3.43, 9.45) | 0.374 |
| WBC, ×109/L | 18.31 (14.72, 21.83) | 17.25 (14.17, 21.66) | 19.04 (15.97, 22.78) | 0.110 | 17.25 (14.15, 21.66) | 19.10 (16.34, 22.88) | 0.038 |
| Hemoglobin, g/L | 121.00 (100.75, 135.00) | 119.00 (100.75, 132.25) | 126.50 (98.50, 138.50) | 0.192 | 120.50 (101.75, 133.00) | 126.50 (91.25, 138.50) | 0.653 |
| Platelet, ×109/L | 219.00 (169.00, 273.50) | 215.50 (168.75, 265.50) | 233.00 (167.00, 285.75) | 0.240 | 215.50 (168.75, 264.25) | 215.50 (167.00, 305.50) | 0.099 |
| Albumin, g/L | 34.90 ± 5.07 | 34.92 ± 4.93 | 34.86 ± 5.41 | 0.946 | 35.12 ± 4.89 | 34.24 ± 5.58 | 0.369 |
| Creatinine, μmol/L | 122.00 (91.00, 168.25) | 110.00 (84.75, 143.50) | 153.10 (117.50, 224.75) | <0.001 | 111.50 (84.75, 143.25) | 164.50 (128.25, 256.25) | <0.001 |
| UA, μmol/L | 452.00 (361.00, 561.75) | 381.00 (314.25, 456.75) | 638.50 (577.75, 719.00) | <0.001 | 392.50 (327.75, 477.50) | 664.50 (600.75, 777.00) | <0.001 |
| Potassium, mmol/L | 4.36 (3.98, 4.66) | 4.29 (3.95, 4.63) | 4.46 (4.04, 5.08) | 0.112 | 4.31 (3.95, 4.65) | 4.40 (4.04, 5.09) | 0.234 |
| Sodium, mmol/L | 139.00 (136.00, 141.53) | 139.15 (136.60, 141.50) | 138.00 (135.00, 142.53) | 0.387 | 139.00 (136.58, 141.50) | 138.50 (134.93, 142.53) | 0.476 |
| Calcium, mmol/L | 2.08 ± 0.19 | 2.07 ± 0.18 | 2.09 ± 0.21 | 0.545 | 2.08 ± 0.19 | 2.07 ± 0.21 | 0.841 |
| Glucose, mmol/L | 13.22 (10.43, 18.28) | 13.45 (10.43, 18.06) | 12.90 (10.40, 20.26) | 0.860 | 13.45 (10.43, 18.06) | 12.90 (9.98, 20.92) | 0.855 |
| Total cholesterol, mmol/L | 3.89 (3.16, 5.05) | 3.79 (3.13, 5.00) | 4.29 (3.17, 5.10) | 0.132 | 3.79 (3.14, 5.00) | 4.65 (3.17, 5.10) | 0.100 |
| Triglycerides, mmol/L | 1.17 (0.86, 1.85) | 1.09 (0.78, 1.73) | 1.40 (0.96, 1.86) | 0.048 | 1.10 (0.80, 1.79) | 1.40 (0.96, 1.86) | 0.145 |
| LDL-C, mmol/L | 2.36 (1.77, 3.28) | 2.21 (1.73, 3.21) | 2.50 (1.87, 3.51) | 0.218 | 2.21 (1.73, 3.21) | 2.53 (1.97, 3.51) | 0.222 |
| HDL-C, mmol/L | 1.00 (0.78, 1.23) | 1.04 (0.80, 1.26) | 0.91 (0.73, 1.19) | 0.148 | 1.02 (0.80, 1.25) | 0.91 (0.73, 1.23) | 0.286 |
| Fibrinogen, g/L | 3.44 (2.52, 4.77) | 3.48 (2.72, 4.82) | 3.36 (2.38, 4.63) | 0.354 | 3.30 (2.55, 4.82) | 3.69 (2.45, 4.63) | 0.909 |
| cTnT, ng/mL | 10.00 (4.23, 10.00) | 9.16 (4.26, 10.00) | 10.00 (4.10, 10.00) | 0.643 | 9.05 (4.26, 10.00) | 10.00 (4.10, 10.00) | 0.521 |
| CK-MB, ng/mL | 212.10 (48.79, 300.00) | 182.25 (44.04, 300.00) | 251.50 (65.42, 300.00) | 0.568 | 204.25 (44.04, 300.00) | 229.65 (65.42, 300.00) | 0.812 |
| Nt-ProBNP, pg/mL | 4,709.50 (1,854.00, 11,167.00) | 4,287.00 (1,875.75, 9,279.25) | 6,868.50 (1,652.75, 12,344.25) | 0.341 | 4,090.00 (1,778.75, 9,098.50) | 7,170.00 (1,881.50, 21,521.25) | 0.083 |
| ECMO timing, n (%) | 0.097 | 0.093 | |||||
| Before revascularization | 103 (72.5) | 67 (68.4) | 36 (81.8) | 73 (68.9) | 30 (83.3) | ||
| After revascularization | 39 (27.5) | 31 (31.6) | 8 (18.2) | 33 (31.1) | 6 (16.7) | ||
| ECMO initiation location, n (%) | 0.874 | 0.625 | |||||
| In-hospital | 102 (71.8) | 70 (71.4) | 32 (72.7) | 75 (70.8) | 27 (75.0) | ||
| Out-of-hospital | 40 (28.2) | 28 (28.6) | 12 (27.3) | 31 (29.2) | 9 (25.0) | ||
| IABP, n (%) | 61 (43.0) | 40 (40.8) | 21 (47.7) | 0.442 | 43 (40.6) | 18 (50.0) | 0.323 |
| ECMO duration, days | 9.00 (6.00, 12.00) | 9.00 (6.00, 12.00) | 8.00 (3.25, 11.00) | 0.113 | 8.50 (6.00, 12.00) | 9.00 (3.25, 11.75) | 0.670 |
| Mechanical ventilation duration, days | 1.50 (1.00, 6.00) | 2.00 (1.00, 6.25) | 1.00 (1.00, 5.00) | 0.435 | 2.00 (1.00, 6.25) | 1.00 (0.25, 4.75) | 0.294 |
| IABP duration, days | 0.00 (0.00, 9.00) | 0.00 (0.00, 8.25) | 0.00 (0.00, 9.00) | 0.454 | 0.00 (0.00, 8.25) | 1.00 (0.00, 11.25) | 0.281 |
| CRRT duration, days | 0.00 (0.00, 0.00) | 0.00 (0.00, 0.00) | 0.00 (0.00, 1.00) | 0.091 | 0.00 (0.00, 0.00) | 0.00 (0.00, 1.75) | 0.032 |
| 28-day mortality, n (%) | 0.002 | 0.001 | |||||
| Yes | 48 (33.8) | 25 (25.5) | 23 (52.3) | 28 (26.4) | 20 (55.6) | ||
| No | 94 (66.2) | 73 (74.5) | 21 (47.7) | 78 (73.6) | 16 (44.4) | ||
| 90-day mortality, n (%) | 0.009 | 0.004 | |||||
| Yes | 58 (40.8) | 33 (33.7) | 25 (56.8) | 36 (43.3) | 22 (61.1) | ||
| No | 84 (59.2) | 65 (66.3) | 19 (43.2) | 70 (66.0) | 14 (38.9) |
UA, uric acid; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; MI, myocardial infarction; SCAI, Society for Cardiovascular Angiography and Interventions; ECMO, extracorporeal membrane oxygenation; TIMI, Thrombolysis in Myocardial Infarction; LVEF, left ventricular ejection fraction; WBC, white blood cell; cTnT, cardiac troponin T; CK-MB, creatine kinase-MB; Nt-ProBNP, N-terminal pro-brain natriuretic peptide; IABP, intra-aortic balloon pump; CRRT, continuous renal replacement therapy.
Using the exploratory ROC-derived cut-off for 90-day mortality (561 μmol/L), patients were reclassified into a low-UA group (≤561 μmol/L) and a high-UA group (>561 μmol/L). The median UA values were 392.50 μmol/L (327.75–477.50 μmol/L) in the low-UA group and 664.50 μmol/L (600.75–777.00 μmol/L) in the high-UA group. Under this stratification, the high-UA group remained younger but also displayed lower LVEF, accompanied by higher BMI, WBC count, creatinine level, and longer CRRT duration (all P < 0.05). Consistently, 28-day and 90-day mortality were increased among patients with UA > 561 μmol/L (P = 0.001 and 0.004, respectively), whereas the remaining clinical variables were comparable between groups (P > 0.05).
3.2. Clinical characteristics according to 28-day and 90-day mortality
Table 2 presented clinical features stratified by survival status. When grouped by 28-day outcome (death, n = 48; survival, n = 94), patients who died within 28 days demonstrated lower LVEF and a reduced prevalence of TIMI grade 3, while exhibiting higher WBC count, creatinine, and NT-proBNP, together with prolonged mechanical ventilation, longer CRRT duration, and higher CRRT intensity/level (all P < 0.05).
Table 2.
Clinical characteristics grouped by 28-day and 90-day mortality.
| Variables | 28-day mortality | 90-day mortality | ||||
|---|---|---|---|---|---|---|
| No | Yes | P value | No | Yes | P value | |
| N | 94 | 48 | 84 | 58 | ||
| Age, years | 56.97 ± 11.70 | 58.65 ± 11.21 | 0.414 | 56.18 ± 11.59 | 59.50 ± 11.25 | 0.092 |
| Sex, n (%) | 0.352 | 0.088 | ||||
| Male | 72 (76.6) | 40 (83.3) | 64 (76.2) | 48 (82.8) | ||
| Female | 22 (23.4) | 8 (16.7) | 20 (23.8) | 10 (17.2) | ||
| Smoking, n (%) | 43 (45.7) | 23 (47.9) | 0.806 | 39 (46.4) | 27 (46.6) | 0.988 |
| Hypertension, n (%) | 31 (33.0) | 22 (45.8) | 0.134 | 28 (33.3) | 25 (43.1) | 0.237 |
| Diabetes, n (%) | 30 (31.9) | 15 (31.3) | 0.936 | 27 (32.1) | 18 (31.0) | 0.889 |
| Previous MI, n (%) | 9 (9.6) | 9 (18.8) | 0.120 | 8 (10.6) | 10 (17.2) | 0.174 |
| SCAI shock state, n (%) | 0.781 | 0.726 | ||||
| C | 11 (11.7) | 5 (10.4) | 10 (11.9) | 6 (10.3) | ||
| D | 51 (54.3) | 29 (60.4) | 45 (53.6) | 35 (60.3) | ||
| E | 32 (34.0) | 14 (29.2) | 29 (34.5) | 17 (29.3) | ||
| Resuscitation before ECMO, n (%) | 36 (38.3) | 17 (35.4) | 0.737 | 35 (41.7) | 18 (31.0) | 0.918 |
| BMI, kg/m2 | 25.06 (22.81, 27.55) | 24.22 (22.62, 27.00) | 0.301 | 25.31 (22.86, 27.62) | 24.22 (22.39, 26.26) | 0.049 |
| SBP, mmHg | 72.50 (0.00, 86.00) | 68.00 (0.00, 80.00) | 0.301 | 71.00 (0.00, 86.00) | 70.00 (37.50, 83.50) | 0.910 |
| DBP, mmHg | 45.50 (0.00, 60.00) | 45.00 (0.00, 60.00) | 0.568 | 41.00 (0.00, 57.50) | 47.00 (22.50, 59.00) | 0.422 |
| HR, beats/min | 68.50 (0.00, 103.00) | 88.50 (0.00, 110.00) | 0.467 | 65.50 (0.00, 102.00) | 90.00 (22.50, 110.00) | 0.095 |
| Diseased vessels ≥2, n (%) | 55 (58.5) | 29 (60.4) | 0.827 | 48 (57.1) | 36 (62.1) | 0.557 |
| Culprit vessels ≥2, n (%) | 31 (33.0) | 21 (43.8) | 0.208 | 27 (32.1) | 25 (43.1) | 0.183 |
| Stents implanted ≥2, n (%) | 17 (18.5) | 11 (22.9) | 0.494 | 15 (17.9) | 13 (22.4) | 0.502 |
| TIMI grade, n (%) | 0.001 | 0.005 | ||||
| 0 | 3 (3.2) | 5 (10.4) | 2 (2.4) | 6 (10.3) | ||
| 1 | 0 (0.0) | 1 (2.1) | 0 (0.0) | 1 (1.7) | ||
| 2 | 4 (4.3) | 9 (18.8) | 4 (4.8) | 9 (15.5) | ||
| 3 | 87 (92.6) | 33 (68.8) | 78 (92.9) | 42 (72.4) | ||
| Time to revascularization, h | 7.00 (4.00, 16.00) | 7.00 (4.00, 48.00) | 0.423 | 6.50 (4.00, 19.75) | 7.25 (4.00, 40.50) | 0.682 |
| LVEF, % | 36.21 ± 9.51 | 29.42 ± 8.27 | <0.001 | 37.06 ± 9.32 | 29.36 ± 8.21 | <0.001 |
| Lactate, mmol/L | 5.00 (3.28, 7.50) | 6.35 (3.43, 9.68) | 0.071 | 4.80 (3.05, 7.48) | 6.10 (3.65, 9.43) | 0.018 |
| WBC, ×109/L | 16.99 (13.87, 21.66) | 19.35 (15.97, 22.62) | 0.025 | 16.67 (12.90, 22.10) | 19.00 (15.83, 21.65) | 0.020 |
| Hemoglobin, g/L | 120.50 (99.50, 133.50) | 122.00 (101.50, 135.00) | 0.988 | 123.00 (100.50, 135.00) | 118.00 (100.75, 134.25) | 0.554 |
| Platelet, ×109/L | 213.50 (165.75, 273.50) | 228.00 (176.25, 278.50) | 0.600 | 212.50 (164.25, 276.50) | 224.50 (186.50, 264.25) | 0.706 |
| Albumin, g/L | 35.31 ± 5.12 | 34.09 ± 4.92 | 0.174 | 35.38 ± 5.24 | 34.20 ± 4.77 | 0.173 |
| Creatinine, μmol/L | 113.00 (83.00, 145.00) | 149.10 (109.75, 202.75) | <0.001 | 109.00 (80.00, 142.75) | 143.00 (111.25, 202.25) | <0.001 |
| Potassium, mmol/L | 4.34 (3.94, 4.64) | 4.40 (4.03, 4.80) | 0.309 | 4.28 (3.92, 4.63) | 4.41 (4.05, 4.81) | 0.159 |
| Sodium, mmol/L | 139.00 (136.00, 141.28) | 138.75 (135.68, 142.53) | 0.682 | 139.25 (136.05, 141.43) | 138.75 (135.60, 142.18) | 0.495 |
| Calcium, mmol/L | 2.08 ± 0.20 | 2.07 ± 0.17 | 0.667 | 2.09 ± 0.21 | 2.07 ± 0.17 | 0.550 |
| Glucose, mmol/L | 13.22 (10.15, 18.06) | 13.25 (11.23, 20.43) | 0.277 | 13.22 (10.16, 18.13) | 13.25 (10.88, 20.42) | 0.407 |
| Total cholesterol, mmol/L | 3.90 (3.20, 5.14) | 3.86 (3.09, 4.77) | 0.725 | 3.92 (3.17, 5.13) | 3.83 (3.14, 4.83) | 0.648 |
| Triglycerides, mmol/L | 1.16 (0.85, 1.91) | 1.21 (0.86, 1.65) | 0.481 | 1.23 (0.87, 1.92) | 1.14 (0.83, 1.62) | 0.190 |
| LDL-C, mmol/L | 2.42 (1.85, 3.55) | 2.33 (1.65, 3.04) | 0.150 | 2.42 (1.86, 3.52) | 2.33 (1.66, 3.06) | 0.170 |
| HDL-C, mmol/L | 0.98 (0.78, 1.18) | 1.06 (0.76, 1.29) | 0.349 | 0.98 (0.77, 1.17) | 1.06 (0.78, 1.30) | 0.207 |
| Fibrinogen, g/L | 3.24 (2.43, 4.83) | 3.67 (2.73, 4.51) | 0.652 | 3.21 (2.39, 4.81) | 3.67 (2.76, 4.62) | 0.374 |
| cTnT, ng/mL | 8.62 (4.30, 10.00) | 10.00 (3.84, 10.00) | 0.366 | 8.28 (4.42, 10.00) | 10.00 (3.95, 10.00) | 0.311 |
| CK-MB, ng/mL | 182.25 (28.68, 300.00) | 278.60 (106.63, 300.00) | 0.129 | 177.55 (27.75, 300.00) | 278.60 (108.88, 300.00) | 0.077 |
| Nt-ProBNP, pg/mL | 3,402.50 (1,357.25, 8,052.75) | 6,995.50 (2,373.63, 15,470.00) | 0.008 | 3,039.00 (1,264.50, 7,456.50) | 6,995.50 (2,413.88, 20,622.75) | 0.001 |
| ECMO timing, n (%) | 0.206 | 0.262 | ||||
| Before revascularization | 65 (69.1) | 38 (79.2) | 58 (69.0) | 45 (77.6) | ||
| After revascularization | 29 (30.9) | 10 (20.8) | 26 (31.0) | 13 (22.4) | ||
| ECMO initiation location, n (%) | 0.165 | 0.375 | ||||
| In-hospital | 64 (68.1) | 38 (79.2) | 58 (69.0) | 44 (75.9) | ||
| Out-of-hospital | 30 (31.9) | 10 (20.8) | 26 (31.0) | 14 (24.1) | ||
| IABP, n (%) | 37 (39.4) | 24 (50.0) | 0.226 | 30 (35.7) | 31 (53.4) | 0.036 |
| ECMO duration, days | 9.00 (6.00, 12.00) | 8.00 (4.00, 11.75) | 0.409 | 8.00 (6.00, 11.75) | 9.00 (5.50, 12.25) | 0.988 |
| Mechanical ventilation duration, days | 1.00 (0.00, 4.25) | 4.50 (1.00, 8.75) | 0.001 | 1.00 (0.00, 4.00) | 4.00 (1.00, 8.25) | 0.003 |
| IABP duration, days | 0.00 (0.00, 9.25) | 0.00 (0.00, 6.75) | 0.861 | 0.00 (0.00, 9.00) | 1.00 (0.00, 8.25) | 0.210 |
| CRRT duration, days | 0.00 (0.00, 0.00) | 0.00 (0.00, 2.00) | <0.001 | 0.00 (0.00, 0.00) | 0.00 (0.00, 2.00) | <0.001 |
| UA, μmol/L | 433.50 (345.25, 525.25) | 516.50 (380.75, 608.75) | 0.011 | 433.50 (327.75, 525.75) | 488.50 (380.00, 598.25) | 0.015 |
| Standardized UA | −0.24 (−0.72, 0.27) | 0.22 (−0.53, 0.73) | 0.011 | −0.24 (−0.82, 0.27) | 0.07 (−0.53, 0.68) | 0.015 |
| UA categorical variable, n (%) | 0.002 | 0.009 | ||||
| Low UA (≤530) | 73 (77.7) | 25 (52.1) | 65 (77.4) | 33 (56.9) | ||
| High UA (>530) | 21 (22.3) | 23 (47.9) | 19 (22.6) | 25 (43.1) | ||
| UA categorical variable, n (%) | 0.001 | 0.004 | ||||
| Low UA (≤561) | 78 (83.0) | 28 (58.3) | 70 (83.3) | 36 (62.1) | ||
| High UA (>561) | 16 (17.0) | 20 (41.7) | 14 (21.3) | 22 (37.9) | ||
UA, uric acid; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; MI, myocardial infarction; SCAI, Society for Cardiovascular Angiography and Interventions; ECMO, extracorporeal membrane oxygenation; TIMI, Thrombolysis in Myocardial Infarction; LVEF, left ventricular ejection fraction; WBC, white blood cell; cTnT, cardiac troponin T; CK-MB, creatine kinase-MB; Nt-ProBNP, N-terminal pro-brain natriuretic peptide; IABP, intra-aortic balloon pump; CRRT, continuous renal replacement therapy.
A broadly concordant pattern was observed for 90-day mortality: patients who died by 90 days showed lower LVEF and higher lactate, WBC, creatinine, NT-proBNP, increased IABP utilization, longer durations of mechanical ventilation and CRRT, and higher UA levels (all P < 0.05). No statistically significant differences were detected for the remaining variables (P > 0.05).
3.3. Univariable Cox regression for 28-day and 90-day mortality
As shown in Table 3, univariable Cox regression identified TIMI grade, LVEF, lactate, creatinine, NT-proBNP, and UA as significant correlates of both 28-day and 90-day mortality (all P < 0.05). Quantitatively, each 1 μmol/L increment in UA was associated with a 0.2% increase in the hazard of death at 28 days and 90 days (HR: 1.002, 95% CI: 1.001–1.004, P = 0.003; HR: 1.002, 95% CI: 1.001–1.003, P = 0.001). When scaled per 1-SD increase, UA was associated with a 47.1% and 47.8% elevation in the hazard of 28-day and 90-day mortality, respectively (HR: 1.471, 95% CI: 1.137–1.905, P = 0.003; HR: 1.478, 95% CI: 1.164–1.877, P = 0.001).
Table 3.
Univariate Cox regression analysis for 28-day and 90-day mortality.
| Variables | 28-day mortality | 90-day mortality | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Age | 1.013 | 0.988–1.038 | 0.324 | 1.020 | 0.997–1.040 | 0.087 |
| Sex | ||||||
| Male | 1.401 | 0.656–2.994 | 0.384 | 1.363 | 0.690–2.694 | 0.373 |
| Female | Ref | Ref | ||||
| Smoking | 1.068 | 0.606–1.881 | 0.820 | 1.011 | 0.604–1.695 | 0.695 |
| Hypertension | 1.549 | 0.877–2.734 | 0.131 | 1.393 | 0.828–2.344 | 0.211 |
| Diabetes | 0.960 | 0.521–1.767 | 0.896 | 0.948 | 0.544–1.654 | 0.852 |
| Previous MI | 1.927 | 0.932–3.986 | 0.077 | 1.790 | 0.904–3.543 | 0.095 |
| SCAI Shock state | ||||||
| C | Ref | Ref | ||||
| D | 1.237 | 0.479–3.196 | 0.661 | 1.249 | 0.525–2.971 | 0.614 |
| E | 1.010 | 0.364–2.804 | 0.985 | 1.016 | 0.400–2.576 | 0.974 |
| Resuscitation before ECMO | 0.931 | 0.515–1.681 | 0.812 | 0.751 | 0.431–1.311 | 0.314 |
| BMI, kg/m2 | 0.941 | 0.862–1.027 | 0.173 | 0.917 | 0.845–0.995 | 0.038 |
| SBP, mmHg | 0.998 | 0.990–1.006 | 0.626 | 1.001 | 0.994–1.008 | 0.788 |
| DBP, mmHg | 0.998 | 0.987–1.009 | 0.742 | 1.003 | 0.993–1.014 | 0.561 |
| HR, beats/min | 1.002 | 0.996–1.007 | 0.587 | 1.003 | 0.998–1.009 | 0.197 |
| Diseased vessels ≥2 | 1.014 | 0.568–1.808 | 0.963 | 1.097 | 0.645–1.865 | 0.733 |
| Culprit vessels ≥2 | 1.364 | 0.771–2.413 | 0.286 | 1.350 | 0.802–2.270 | 0.258 |
| Stents implanted ≥2 | 1.164 | 0.594–2.281 | 0.659 | 1.136 | 0.613–2.107 | 0.685 |
| TIMI grade | ||||||
| 0 | Ref | Ref | ||||
| 1 | 1.819 | 0.208–15.864 | 0.588 | 1.713 | 0.202–14.525 | 0.621 |
| 2 | 0.917 | 0.307–2.743 | 0.877 | 0.774 | 0.275–2.177 | 0.627 |
| 3 | 0.241 | 0.094–0.619 | 0.003 | 0.248 | 0.105–0.585 | 0.001 |
| Time to revascularization | 1.001 | 0.999–1.003 | 0.344 | 1.001 | 0.998–1.003 | 0.593 |
| LVEF | 0.932 | 0.900–0.964 | <0.001 | 0.927 | 0.898–0.957 | <0.001 |
| Lactate | 1.083 | 1.013–1.158 | 0.019 | 1.091 | 1.027–1.160 | 0.005 |
| WBC | 1.034 | 0.997–1.072 | 0.073 | 1.032 | 0.998–1.067 | 0.069 |
| Hemoglobin | 1.000 | 0.988–1.012 | 0.986 | 0.998 | 0.987–1.008 | 0.690 |
| Platelet | 0.999 | 0.997–1.002 | 0.620 | 0.999 | 0.997–1.002 | 0.476 |
| Albumin | 0.966 | 0.917–1.018 | 0.201 | 0.967 | 0.921–1.015 | 0.173 |
| Creatinine | 1.003 | 1.001–1.005 | <0.001 | 1.003 | 1.002–1.005 | <0.001 |
| Potassium | 1.205 | 0.831–1.748 | 0.326 | 1.252 | 0.897–1.746 | 0.186 |
| Sodium | 0.976 | 0.912–1.004 | 0.477 | 0.961 | 0.902–1.024 | 0.218 |
| Calcium | 0.748 | 0.176–3.187 | 0.695 | 0.700 | 0.188–2.606 | 0.595 |
| Glucose | 1.036 | 0.992–1.083 | 0.109 | 1.032 | 0.991–1.075 | 0.131 |
| Total cholesterol | 1.012 | 0.821–1.248 | 0.908 | 1.003 | 0.828–1.214 | 0.979 |
| Triglycerides | 0.814 | 0.565–1.171 | 0.267 | 0.791 | 0.565–1.107 | 0.172 |
| LDL-C | 0.832 | 0.637–1.085 | 0.175 | 0.859 | 0.677–1.089 | 0.210 |
| HDL-C | 1.411 | 0.782–2.546 | 0.253 | 1.649 | 1.000–2.718 | 0.050 |
| Fibrinogen | 1.043 | 0.878–1.240 | 0.630 | 1.069 | 0.916–1.248 | 0.396 |
| cTnT | 1.035 | 0.949–1.127 | 0.439 | 1.032 | 0.955–1.116 | 0.427 |
| CK-MB | 1.002 | 0.999–1.004 | 0.154 | 1.002 | 1.000–1.004 | 0.092 |
| Nt-ProBNP | 1.000 | 1.000–1.000 | 0.010 | 1.000 | 1.000–1.000 | <0.001 |
| ECMO timing | ||||||
| Before revascularization | Ref | Ref | ||||
| After revascularization | 0.654 | 0.326–1.312 | 0.231 | 0.711 | 0.384–1.318 | 0.279 |
| ECMO initiation location | ||||||
| In-hospital | Ref | Ref | ||||
| Out-of-hospital | 0.624 | 0.311–1.252 | 0.184 | 0.745 | 0.408–1.359 | 0.336 |
| IABP | 1.382 | 0.785–2.435 | 0.262 | 1.623 | 0.968–2.720 | 0.066 |
| ECMO duration | 0.998 | 0.961–1.037 | 0.921 | 1.006 | 0.976–1.038 | 0.681 |
| Mechanical ventilation duration | 1.026 | 0.996–1.058 | 0.092 | 1.027 | 0.999–1.056 | 0.062 |
| IABP duration | 0.986 | 0.951–1.022 | 0.427 | 1.002 | 0.973–1.031 | 0.919 |
| CRRT duration | 1.020 | 0.983–1.058 | 0.301 | 1.025 | 0.993–1.058 | 0.132 |
| UA | 1.002 | 1.001–1.004 | 0.003 | 1.002 | 1.001–1.003 | 0.001 |
| Standardized UA | 1.471 | 1.137–1.905 | 0.003 | 1.478 | 1.164–1.877 | 0.001 |
| UA categorical variable | ||||||
| Low UA (≤530) | Ref | Ref | ||||
| High UA (>530) | 2.592 | 1.470–4.570 | 0.001 | 2.194 | 1.304–3.693 | 0.003 |
| UA categorical variable | ||||||
| Low UA (≤561) | Ref | Ref | ||||
| High UA (>561) | 2.632 | 1.481–4.676 | 0.001 | 2.341 | 1.375–3.984 | 0.002 |
UA, uric acid; BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; HR, heart rate; MI, myocardial infarction; SCAI, Society for Cardiovascular Angiography and Interventions; ECMO, extracorporeal membrane oxygenation; TIMI, Thrombolysis in Myocardial Infarction; LVEF, left ventricular ejection fraction; WBC, white blood cell; cTnT, cardiac troponin T; CK-MB, creatine kinase-MB; Nt-ProBNP, N-terminal pro-brain natriuretic peptide; IABP, intra-aortic balloon pump; CRRT, continuous renal replacement therapy; Ref, reference category.
Using 530 μmol/L as the threshold, high UA conferred a 2.592-fold higher risk of 28-day mortality (95% CI: 1.470–4.570, P = 0.001) and a 2.194-fold higher risk of 90-day mortality (95% CI: 1.304–3.693, P = 0.003) relative to low UA. With 561 μmol/L as the cut-off, the excess risk increased to 2.632-fold for 28-day mortality (95% CI: 1.481–4.676, P = 0.001) and 2.341-fold for 90-day mortality (95% CI: 1.375–3.984, P = 0.002).
3.4. Association between UA and mortality risk at 28 and 90 days
The independent relationship between UA and short- and intermediate-term mortality was further corroborated in multivariable Cox models (Table 4). After adjustment for TIMI 3 vs. TIMI 0–2 and prespecified clinical covariates, each 1 μmol/L increase in UA remained associated with a higher hazard of 28-day mortality (HR: 1.002, 95% CI: 1.000–1.003, P = 0.038), and each 1-SD increase corresponded to a 37.9% increase in risk (HR: 1.379, 95% CI: 1.019–1.867, P = 0.038). Consistently, UA > 530 μmol/L was associated with a 2.499-fold higher risk of 28-day death compared with UA ≤ 530 μmol/L (95% CI: 1.335–4.677, P = 0.004).
Table 4.
Association between UA and mortality.
| Variables | Model 1 | Model 2 | Model 3 | ||||||
|---|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | HR | 95% CI | P value | |
| 28-day mortality | |||||||||
| UA as continuous variable | |||||||||
| UA | 1.002 | 1.001–1.004 | 0.003 | 1.002 | 1.001–1.004 | 0.002 | 1.002 | 1.000–1.003 | 0.038 |
| Standardized UA | 1.471 | 1.137–1.905 | 0.003 | 1.495 | 1.156–1.934 | 0.002 | 1.379 | 1.019–1.867 | 0.038 |
| UA as categorical variable (530) | |||||||||
| Low UA group | Ref | Ref | Ref | ||||||
| High UA group | 2.592 | 1.470–4.570 | 0.001 | 3.004 | 1.684–5.357 | <0.001 | 2.499 | 1.335–4.677 | 0.004 |
| 90-day mortality | |||||||||
| UA as continuous variable | |||||||||
| UA | 1.002 | 1.001–1.003 | 0.001 | 1.002 | 1.001–1.004 | <0.001 | 1.002 | 1.001–1.004 | 0.005 |
| Standardized UA | 1.478 | 1.164–1.877 | 0.001 | 1.504 | 1.184–1.910 | <0.001 | 1.494 | 1.126–1.982 | 0.005 |
| UA as categorical variable (561) | |||||||||
| Low UA group | Ref | Ref | Ref | ||||||
| High UA group | 2.341 | 1.375–3.984 | 0.002 | 2.596 | 1.514–4.450 | <0.001 | 2.262 | 1.222–4.188 | 0.009 |
For 28-day mortality, Model 1: unadjusted; Model 2: adjusted for TIMI 3 vs. 0–2 only; Model 3: adjusted for TIMI 3 vs. 0–2, LVEF, lactate, creatinine, and Nt-ProBNP.
For 90-day mortality, Model 1: unadjusted; Model 2: adjusted for TIMI 3 vs. 0–2 only; Model 3: adjusted for TIMI 3 vs. 0–2, BMI, LVEF, lactate, creatinine, and Nt-ProBNP.
UA, uric acid; BMI, body mass index; TIMI, Thrombolysis in Myocardial Infarction; LVEF, left ventricular ejection fraction; Nt-ProBNP, N-terminal pro-brain natriuretic peptide; Ref, reference category; HR, hazard ratio; CI, confidence interval.
For 90-day mortality, UA was likewise associated with outcome: each 1 μmol/L increase was linked to a higher hazard (HR: 1.002, 95% CI: 1.001–1.004, P = 0.005), while each 1-SD increase corresponded to a 49.4% elevation in 90-day mortality risk (HR: 1.494, 95% CI: 1.126–1.982, P = 0.005). Patients with UA > 561 μmol/L had a 2.262-fold higher risk of 90-day mortality than those with UA ≤ 561 μmol/L (95% CI: 1.222–4.188, P = 0.009).
Considering the significance of IABP in VA-ECMO, an additional subgroup analysis based on IABP was conducted (Supplementary Table S1). The results showed that in the subgroup without IABP, for every 1 SD increase in UA, the 28-day and 90-day mortality risks increased by 49.5% and 58.2% respectively (P < 0.05). Moreover, the 28-day and 90-day mortality risks in the high UA group were 2.403 times and 3.003 times those of the low UA group, respectively (P < 0.05). However, in the subgroup with IABP, only the high UA group had a higher 28-day mortality risk (P = 0.016), while the others showed no statistical significance.
3.5. Visualization of the prognostic performance of UA
As depicted in Figure 1, ROC analysis was performed as an exploratory assessment of UA discrimination for 28-day and 90-day mortality. In the original cohort, the apparent AUCs were 0.631 (95% CI: 0.532–0.728) and 0.621 (95% CI: 0.523–0.715), respectively, with apparent Youden-derived cut-offs of 530 μmol/L and 561 μmol/L. Bootstrap internal validation showed modest discrimination and sampling variability in the re-estimated cut-offs. The bootstrap median cut-off was 534 μmol/L (371–586 μmol/L) for 28-day mortality and 530 μmol/L (300–571 μmol/L) for 90-day mortality (Supplementary Table S2).
Figure 1.
ROC curves for UA in predicting 28-day (A) and 90-day mortality (B) UA, uric acid; ROC, receiver operating characteristic; AUC, area under the curve; CI, confidence interval.
Kaplan–Meier curves further demonstrated pronounced divergence in cumulative mortality risk between UA strata (Figure 2). For the 28-day endpoint, patients with UA > 530 μmol/L exhibited higher cumulative mortality than those with UA ≤ 530 μmol/L (log-rank P = 0.001; Figure 2A). Similarly, for the 90-day endpoint, UA >561 μmol/L was associated with lower survival rate compared with UA ≤ 561 μmol/L (log-rank P = 0.001; Figure 2B).
Figure 2.
Kaplan–Meier survival curves showing the differences in cumulative risks of 28-day mortality (A) and 90-day mortality (B) among different UA groups. UA, uric acid.
RCS analyses supported a statistically linear association between UA and mortality risk at both 28 days (P = 0.031; P for non-linearity = 0.758) and 90 days (P = 0.018; P for non-linearity = 0.936), with no evidence of a non-linear relationship (Figure 3).
Figure 3.
RCS plots showing the correlation between UA and 28-day mortality (A) and 90-day mortality (B) UA, uric acid; RCS, restricted cubic spline.
4. Discussion
Leveraging real-world clinical data, this study applied a suite of complementary statistical approaches to interrogate, from multiple perspectives, the relationship between UA and 28-day as well as 90-day mortality after VA-ECMO support in patients with AMI-CS. After adjustment for prespecified confounders, elevated UA remained associated with increased risks of death at both time points. ROC analyses suggested only modest apparent discrimination. Kaplan–Meier curves showed separation across exploratory UA strata, and RCS modeling was compatible with a positive linear association between UA and mortality risk. These findings suggest that UA may provide prognostic information in this cohort, but external validation is needed.
UA, the end product of purine metabolism, exerts antioxidant effects in vivo and has been implicated in neuroprotection; however, its role in cardiovascular disease has remained contentious. In recent years, accumulating evidence has increasingly supported an association between higher UA levels and cardiovascular burden as well as adverse outcomes. In a large, contemporary cohort of 5,888 patients with successfully revascularized AMI, Kim et al. (9) showed that higher on-admission serum UA was associated with increased long-term all-cause mortality and improved risk discrimination when added to traditional prognostic factors. A cohort analysis (16) including 21,386 participants further demonstrated that UA is an independent risk factor for both incident HF and fatal HF: participants with UA < 5.34 mg/dL exhibited a 2.5% incidence of all-cause HF compared with 4.8% among those above this threshold, and the incidence of fatal HF was 1.6% for UA < 4.89 mg/dL vs. 3.3% for higher levels. These findings suggest that increasing UA may be associated with HF onset and prognosis. Consistently, a retrospective cohort study (17) of 757 patients with AMI and HF with preserved ejection fraction showed that higher UA was independently associated with long-term all-cause mortality; after a median follow-up of 4.8 years, mortality in the high-UA group (32.9%) was higher than that in the normouricemic group (15.5%). Moreover, a meta-analysis (18) encompassing 40 cohort studies and 105,609 patients with acute coronary syndrome (ACS) quantitatively consolidated the prognostic relevance of UA, revealing increased risks of all-cause mortality, cardiovascular mortality, and HF among patients with elevated UA. Notably, that meta-analysis also suggested a non-linear dose–response relationship between UA and mortality, raising the possibility that UA may not merely be a surrogate marker but could also contribute directly to pathogenesis. Collectively, these studies support the notion that UA can be considered an independent risk factor for adverse prognosis in AMI, particularly among patients complicated by HF. In addition, Miao et al. (19) reported a positive relationship between UA and adverse events in HF, with each 1 mg/dL increase in UA associated with a 4% rise in all-cause mortality and a 9% increase in the composite endpoint of death or cardiac events. Deis et al. (20) further observed that, in advanced HF, elevated UA correlated with more profound hemodynamic impairment and predicted long-term all-cause mortality as well as composite outcomes.
Compared with previous studies, our current research made some additional contributions in several respects. First, we focused on a specific and clinically heterogeneous cohort of patients with AMI-CS receiving VA-ECMO support. Second, we incorporated multiple conventional cardiovascular risk factors and key clinical variables, constructing multivariable models to mitigate confounding bias. Third, we examined the dose-response pattern between UA and both 28-day and 90-day mortality. Fourth, we we examined metabolic derangements in the context of mechanical circulatory support, highlighting that once hemodynamics are partially restored, metabolic indicators may retain prognostic relevance.
Despite these observations, the mechanistic pathways linking UA to short- and mid-term mortality in this critically ill cohort remain incompletely defined. Existing evidence suggests that elevated UA may contribute to cardiovascular injury through convergent mechanisms, including induction of oxidative stress, activation of inflammatory cascades, promotion of endothelial dysfunction, suppression of nitric oxide bioavailability, and activation of the RAAS (21–23). UA has also been reported to upregulate platelet-derived growth factor, thereby facilitating vascular smooth muscle cell proliferation and accelerating atherosclerotic progression (24). In parallel, hyperuricemia may precipitate or exacerbate metabolic syndrome (25), which is linked to subclinical cardiovascular injury (6). In AMI-CS supported by VA-ECMO, however, elevated UA may also reflect systemic hypoperfusion, oxygen debt, cellular injury, and impaired renal clearance rather than UA-specific biological effects alone. Of particular note, a UK Biobank analysis (26) of nearly 470,000 individuals suggested that the association between higher UA and sudden cardiac death may be mediated predominantly through renal injury, implicating kidney dysfunction as a critical node in UA-related pathogenic circuitry. In our cohort, patients in the high-UA group exhibited higher serum creatinine and longer durations of CRRT, lending clinical support to this hypothesis. Nevertheless, definitive mechanistic inference will require further experimental interrogation in cellular and animal models.
Several limitations warrant consideration. First, as a single-center retrospective analysis with a limited sample size, selection bias cannot be excluded, and external generalizability should be approached with caution. Second, as an observational retrospective cohort without genetic association analyses, our study cannot establish causality between UA and mortality risk. Third, UA was measured only once at baseline; the absence of longitudinal measurements precluded assessment of sustained exposure and temporal trajectories of UA in relation to 28-day and 90-day mortality. Fourth, although patients with documented high-purine load or purine-containing supplementation before baseline UA assessment were excluded and acute nutritional support followed institutional protocols, prehospital dietary purine intake and undocumented purine exposure could not be precisely quantified. Fifth, residual confounding by end-organ perfusion and oxygen delivery remains possible. In AMI-CS, cellular injury and elevated UA may partly reflect systemic hypoperfusion, oxygen debt, and renal dysfunction. Although baseline lactate and creatinine were included as clinically relevant surrogates, single baseline measurements cannot fully capture dynamic lactate clearance, arterial oxygen content, ECMO flow adequacy, microcirculatory perfusion, or the evolving balance between oxygen delivery and metabolic demand. Sixth, although TIMI flow was dichotomized and parsimonious models were used to reduce the degrees of freedom, the number of outcome events remained limited; therefore, model overfitting cannot be completely excluded, and the findings require validation in larger external cohorts. Seventh, available pre-baseline medication records were reviewed, and no urate-lowering therapy was documented before baseline UA measurement. However, other medications that may indirectly influence UA, such as diuretics, sodium-glucose cotransporter 2 inhibitors, or losartan, were not uniformly captured in the structured dataset; undocumented medication exposure could not be entirely excluded. Finally, although bootstrap internal validation was performed, the ROC-derived UA cut-offs were generated from a single-center cohort and remain exploratory. These thresholds should not be interpreted as definitive clinical decision limits, and external validation in larger independent cohorts is required before clinical application.
5. Conclusions
This study showed that higher UA levels were associated with increased 28-day and 90-day mortality among patients with AMI-CS receiving VA-ECMO. RCS analyses suggested a positive linear association between UA and mortality risk. However, because this was a single-center, retrospective, observational cohort study, these findings establish association rather than causality and should be regarded as hypothesis-generating. Further large-scale, prospective, and externally validated studies are needed to determine whether UA provides incremental prognostic value beyond established clinical risk factors and whether it can be incorporated into clinical risk assessment.
Acknowledgments
We thank the ECMO and organ support ward at Linyi People's Hospital for their assistance and support.
Funding Statement
The author(s) declared that financial support was not received for this work and/or its publication.
Footnotes
Edited by: Kaisaierjiang Kadier, First Affiliated Hospital of Xinjiang Medical University, China
Reviewed by: Gerrit J. Noordergraaf, Tilburg NL, Netherlands
Soham Samajpaty, College of Medicine & Sagore Dutta Hospital, India
Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving humans were approved by Ethics Committee of Linyi People's Hospital. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.
Author contributions
RM: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing. XZ: Conceptualization, Methodology, Writing – review & editing. SG: Conceptualization, Methodology, Writing – review & editing. YD: Conceptualization, Methodology, Writing – review & editing. WZ: Conceptualization, Methodology, Writing – review & editing. ZG: Conceptualization, Funding acquisition, Project administration, Supervision, Validation, Writing – review & editing.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm.2026.1839659/full#supplementary-material
References
- 1.Sinha SS, Morrow DA, Kapur NK, Kataria R, Roswell RO. 2025 concise clinical guidance: an ACC expert consensus statement on the evaluation and management of cardiogenic shock: a report of the American College of Cardiology solution set oversight committee. J Am Coll Cardiol. (2025) 85(16):1618–41. 10.1016/j.jacc.2025.02.018 [DOI] [PubMed] [Google Scholar]
- 2.Jung C, Bruno RR, Jumean M, Price S, Krychtiuk KA, Ramanathan K, et al. Management of cardiogenic shock: state-of-the-art. Intensive Care Med. (2024) 50:1814–29. 10.1007/s00134-024-07618-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lorusso R, Shekar K, MacLaren G, Schmidt M, Pellegrino V, Meyns B, et al. ELSO interim guidelines for venoarterial extracorporeal membrane oxygenation in adult cardiac patients. ASAIO J. (2021) 67(8):827–44. 10.1097/MAT.0000000000001510 [DOI] [PubMed] [Google Scholar]
- 4.Geller BJ, Sinha SS, Kapur NK, Bakitas M, Balsam LB, Chikwe J, et al. Escalating and de-escalating temporary mechanical circulatory support in cardiogenic shock: a scientific statement from the American Heart Association. Circulation. (2022) 146(6):e50–68. 10.1161/CIR.0000000000001076 [DOI] [PubMed] [Google Scholar]
- 5.Zeymer U, Freund A, Hochadel M, Ostadal P, Belohlavek J, Rokyta R, et al. Venoarterial extracorporeal membrane oxygenation in patients with infarct-related cardiogenic shock: an individual patient data meta-analysis of randomised trials. Lancet. (2023) 402(10410):1338–46. 10.1016/S0140-6736(23)01607-0 [DOI] [PubMed] [Google Scholar]
- 6.Jakubiak GK, Pawlas N, Starzak M, Stanek A, Cieślar G. Cardiac troponin and subclinical cardiovascular dysfunction in metabolic syndrome: a retrospective analysis based on two diagnostic definitions. BMC Cardiovasc Disord. (2026) 26:143. 10.1186/s12872-026-05520-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gherghina M-E, Peride I, Tiglis M, Neagu TP, Niculae A, Checherita IA. Uric acid and oxidative stress: relationship with cardiovascular, metabolic, and renal impairment. Int J Mol Sci. (2022) 23(6):3188. 10.3390/ijms23063188 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Borghi C, Agnoletti D, Cicero AFG, Lurbe E, Virdis A. Uric acid and hypertension: a review of evidence and future perspectives for the management of cardiovascular risk. Hypertension. (2022) 79(9):1927–36. 10.1161/HYPERTENSIONAHA.122.17956 [DOI] [PubMed] [Google Scholar]
- 9.Kim S, Hwang B-H, Lee KY, Kim CJ, Choo E-H, Lim S, et al. High uric acid levels in acute myocardial infarction provide better long-term prognosis predictive power when combined with traditional risk factors. J Clin Med. (2022) 11(19):5531. 10.3390/jcm11195531 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zuo T, Liu X, Jiang L. Hyperuricemia and coronary heart disease mortality: a meta-analysis of prospective cohort studies. BMC Cardiovasc Disord. (2016) 16(1):207. 10.1186/s12872-016-0382-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Maloberti A, Giannattasio C, Bombelli M, Desideri G, Cicero AFG, Muiesan ML, et al. Hyperuricemia and risk of cardiovascular outcomes: the experience of the URRAH (Uric Acid Right for Heart Health) project. High Blood Press Cardiovasc Prev. (2020) 27(2):121–8. 10.1007/s40292-020-00368-z [DOI] [PubMed] [Google Scholar]
- 12.Cho ER, Brill IK, Gram IT, Brown PE, Jha P. Smoking cessation and short- and longer-term mortality. NEJM Evid. (2024) 3(3):EVIDoa2300272. 10.1056/EVIDoa2300272 [DOI] [PubMed] [Google Scholar]
- 13.Mancia G, Kreutz R, Brunström M, Burnier M, Grassi G, Januszewicz A, et al. 2023 ESH guidelines for the management of arterial hypertension: endorsed by the International Society of Hypertension and the European Renal Association. J Hypertens. (2023) 41(12):1874–2071. 10.1097/HJH.0000000000003480 [DOI] [PubMed] [Google Scholar]
- 14.Bajaj M, McCoy RG, Balapattabi K, Bannuru RR, Bellini NJ, Bennett AK, et al. 2. Diagnosis and classification of diabetes: standards of care in diabetes—2026. Diabetes Care. (2026) 49(Suppl 1):S27–49. 10.2337/dc26-S002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Baran DA, Grines CL, Bailey S, Burkhoff D, Hall SA, Henry TD, et al. SCAI Clinical expert consensus statement on the classification of cardiogenic shock: this document was endorsed by the American College of Cardiology (ACC), the American Heart Association (AHA), the Society of Critical Care Medicine (SCCM), and the Society of Thoracic Surgeons (STS) in April 2019. Catheter Cardiovasc Interv. (2019) 94(1):29–37. 10.1002/ccd.28329 [DOI] [PubMed] [Google Scholar]
- 16.Muiesan ML, Salvetti M, Virdis A, Masi S, Casiglia E, Tikhonoff V, et al. Serum uric acid predicts heart failure in a large Italian cohort: search for a cut-off value in the URic acid Right for heArt Health study. J Hypertens. (2021) 39(1):62–9. 10.1097/HJH.0000000000002589 [DOI] [PubMed] [Google Scholar]
- 17.Kim S, Lee KY, Kim KA, Byeon J, Shin S, Hwang B-H, et al. The role of serum uric acid in survival prediction in patients with acute myocardial infarction accompanied by heart failure with preserved ejection fraction. Front Cardiovasc Med. (2025) 12:1622275. 10.3389/fcvm.2025.1622275 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sun Z, Hui J, Wang Y, Wang J, He R, Lyu L, et al. Serum uric acid level and prognosis of acute coronary syndrome: a systematic review and dose-response meta-analysis. Front Cardiovasc Med. (2026) 12:1670418. 10.3389/fcvm.2025.1670418 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Miao L, Guo M, Pan D, Chen P, Chen Z, Gao J, et al. Serum uric acid and risk of chronic heart failure: a systematic review and meta-analysis. Front Med (Lausanne). (2021) 8:785327. 10.3389/fmed.2021.785327 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Deis T, Rossing K, Ersbøll MK, Wolsk E, Gustafsson F. Uric acid in advanced heart failure: relation to central haemodynamics and outcome. Open Heart. (2022) 9(2):e002092. 10.1136/openhrt-2022-002092 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yu W, Cheng JD. Uric acid and cardiovascular disease: an update from molecular mechanism to clinical perspective. Front Pharmacol. (2020) 11:582680. 10.3389/fphar.2020.582680 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Wei X, Zhang M, Huang S, Lan X, Zheng J, Luo H, et al. Hyperuricemia: a key contributor to endothelial dysfunction in cardiovascular diseases. FASEB J. (2023) 37(7):e23012. 10.1096/fj.202300393R [DOI] [PubMed] [Google Scholar]
- 23.Wang K, Zhang Y, Zhou M, Du Y, Li P, Guan C, et al. HDAC Inhibitors alleviate uric acid-induced vascular endothelial cell injury by way of the HDAC6/FGF21/PI3K/AKT pathway. J Cardiovasc Pharmacol. (2023) 81(2):150–64. 10.1097/FJC.0000000000001372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Du L, Zong Y, Li H, Wang Q, Xie L, Yang B, et al. Hyperuricemia and its related diseases: mechanisms and advances in therapy. Signal Transduct Target Ther. (2024) 9(1):212. 10.1038/s41392-024-01916-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Raya-Cano E, Vaquero-Abellán M, Molina-Luque R, De Pedro-Jiménez D, Molina-Recio G, Romero-Saldaña M. Association between metabolic syndrome and uric acid: a systematic review and meta-analysis. Sci Rep. (2022) 12:18412. 10.1038/s41598-022-22025-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Han Y, Li Z, Yao M, Wu J, Xie T, Zhao H, et al. Uric acid and sudden cardiac arrest: a more significant intermediate role of renal insufficiency than cardiac events. Heart Rhythm. (2026) 23(2):306–15. 10.1016/j.hrthm.2025.08.020 [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
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.