Association of the C-reactive Protein–Albumin–Lymphocyte Index With Long-term All-cause Mortality After Percutaneous Coronary Intervention for Acute Coronary Syndrome: A Retrospective Cohort Study

Technical Summary: Association of the C-reactive Protein–Albumin–Lymphocyte Index With Long-term All-cause Mortality After Percutaneous Coronary Intervention for Acute Coronary Syndrome

Problem Statement
Despite advances in revascularization and secondary prevention, patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (PCI) continue to face substantial long-term mortality risks. Traditional risk stratification tools, such as the GRACE score, primarily focus on anatomical and hemodynamic factors but may not fully capture the combined impact of systemic inflammation, nutritional status, and immune competence on post-procedural outcomes. While composite indices like the neutrophil-to-lymphocyte ratio and prognostic nutritional index have shown prognostic value in cardiovascular populations, the specific utility of the C-reactive protein–albumin–lymphocyte (CALLY) index in ACS patients post-PCI remains unclear. This study aims to evaluate the independent association between baseline CALLY and long-term all-cause mortality in this specific clinical setting.

Methodology
The authors conducted a single-center retrospective cohort study involving 388 adult patients with ACS who underwent PCI between January 2021 and January 2024 at the First Affiliated Hospital of Guizhou University of Traditional Chinese Medicine.

• Exposure: The CALLY index was calculated from admission laboratory measurements using the formula: (Albumin [g/dL] × Lymphocyte count [×10⁹/L]) / C-reactive protein [mg/L].
• Outcome: The primary outcome was all-cause mortality during follow-up (median 26 months).
• Statistical Analysis:
  • Cox proportional hazards models were employed to assess the association between CALLY and mortality.
  • Four sequential models were fitted: unadjusted; adjusted for age and sex; adjusted for age, sex, GRACE score, left ventricular ejection fraction (LVEF), and BMI; and a "parsimonious" model adjusting for age, sex, GRACE score, LVEF, diabetes mellitus, chronic kidney disease (CKD), and multivessel disease.
  • Missing data (e.g., smoking status, stent count) were handled using missForest imputation across five datasets, with estimates pooled using Rubin's rules.
  • Dose–response relationships were evaluated using restricted cubic splines. An exploratory two-piecewise Cox model was used to identify potential inflection points.
  • Sensitivity analyses included winsorizing extreme CALLY values, excluding early deaths (≤5th percentile of event times), and calculating E-values to assess robustness against unmeasured confounding.
  • Subgroup analyses were performed to test for effect modification across clinical strata.

Key Results

• Mortality Rates: During the follow-up period, 62 deaths (16.0%) occurred. Mortality rates decreased significantly across increasing CALLY tertiles: 29.2% in the low tertile (0.030–0.461), 13.3% in the medium tertile (0.470–1.150), and 5.4% in the high tertile (1.152–9.816) (P < 0.001).
• Multivariable Association: In the parsimonious adjusted model, each one-unit increase in CALLY was independently associated with a lower risk of all-cause mortality (Hazard Ratio [HR] 0.552; 95% CI 0.368–0.827; P = 0.004). This corresponds to approximately a 45% reduction in the hazard of death per unit increase. The association remained significant when CALLY was modeled per one-standard-deviation increase (HR 0.432; P = 0.004).
• Dose–Response: Restricted cubic spline analysis indicated a predominantly inverse linear relationship with no significant departure from linearity. However, an exploratory two-piecewise analysis identified a data-driven inflection point at CALLY = 3.117, where the inverse association was stronger below the threshold (HR 0.378) and non-significant above it. The authors note this threshold is exploratory and not clinically validated.
• Subgroup Analysis: The inverse association was generally consistent across subgroups. A nominal interaction was observed by sex (P = 0.023), with a stronger association in females (HR 0.176) than males (HR 0.593), and by the number of diseased vessels (P = 0.045), with the strongest effect in patients with ≥4-vessel disease.
• Robustness: Sensitivity analyses confirmed the robustness of the findings. The association persisted after winsorizing extreme values, excluding early deaths, and adjusting for missing data. The E-value for the point estimate was 3.810, suggesting that an unmeasured confounder would need to be strongly associated with both CALLY and mortality (risk ratio ≥ 2.29) to explain away the observed association.

Significance and Claims
The paper concludes that a higher baseline CALLY index is independently associated with lower long-term all-cause mortality in patients with ACS undergoing PCI. The authors position CALLY as a readily available biomarker that integrates inflammatory, nutritional, and immune parameters, potentially reflecting a dimension of host vulnerability not fully captured by traditional risk scores like GRACE.

However, the authors maintain a modest stance regarding the clinical utility of these findings:

1. Hypothesis-Generating: Due to the single-center, retrospective design and lack of external validation, the study is framed as hypothesis-generating rather than definitive proof.
2. No Incremental Validation: The study did not assess whether CALLY improves predictive performance (e.g., via C-statistics or net reclassification improvement) over established models.
3. Threshold Caution: The identified inflection point (3.117) is explicitly stated as a data-driven exploratory finding that should not be interpreted as a clinically validated cut-off.
4. Future Directions: The authors emphasize that external validation and assessment of incremental prognostic value in larger, prospective multicenter studies are required before CALLY can be recommended for clinical decision-making or risk stratification.