Association between the peak hs-CRP-based CALLY Index and CMR-derived microvascular obstruction in patients with STEMI after primary PCI

Microvascular obstruction (MVO) is a common manifestation of impaired microvascular reperfusion in patients with ST-segment elevation myocardial infarction (STEMI) after primary percutaneous coronary intervention (pPCI). It is associated with larger infarct size, impaired left ventricular function, and adverse clinical outcomes. Inflammation, immune dysregulation, and nutritional status are all involved in the pathogenesis and progression of MVO. The CALLY index integrates C-reactive protein, albumin, and lymphocyte count to reflect systemic inflammation, immune status, and nutritional status. However, the association between the CALLY index and CMR-derived MVO in patients with STEMI after pPCI remains unclear. This study aimed to investigate this association. This retrospective study included 524 patients with STEMI who underwent pPCI and in-hospital CMR between January 2022 and June 2025 and met the predefined eligibility criteria. Patients were classified into MVO and non-MVO groups according to CMR-defined MVO status. Clinical, laboratory, and imaging characteristics were compared between the two groups. Among patients with MVO, the association between the CALLY index and MVO extent was assessed using Spearman correlation analysis. Variables with a P value < 0.05 in the univariable logistic regression analyses were considered candidate covariates for the multivariable logistic regression model. Serum albumin, lymphocyte count, and peak hs-CRP were not entered together with the CALLY index because they are components of the index. Peak hs-TnT and peak NT-proBNP were modeled per 1000-unit increase. Exploratory receiver operating characteristic analyses were performed to evaluate the discriminatory performance of the CALLY index and its incremental value when added to the reference model. Continuous net reclassification improvement and integrated discrimination improvement were also calculated. MVO was identified in 48.28% (253/524) of the cohort. Patients with MVO had a significantly lower median CALLY index than those without MVO [1.430 (IQR: 0.631–2.561) vs. 3.200 (IQR: 1.310–7.959), P < 0.001]. Patients with MVO also had higher WBC and neutrophil counts, higher peak hs-CRP, peak hs-TnT, and peak NT-proBNP levels, a larger infarct size (LGE%), and a lower LVEF. Spearman analysis revealed an inverse correlation between CALLY and MVO extent (ρ = − 0.425, P < 0.001 ). Multivariable logistic regression showed that CALLY was independently associated with MVO (OR: 0.81, 95% CI 0.75–0.89, P < 0.001 ). In an exploratory ROC analysis, the CALLY index showed moderate discrimination for MVO, with an AUC of 0.712. Adding the CALLY index to the reference model resulted in a modest numerical increase in the AUC from 0.817 to 0.841 and was accompanied by exploratory improvements in NRI and IDI. A lower CALLY index was independently associated with the presence of CMR-derived MVO in STEMI patients after pPCI and was inversely correlated with MVO extent among patients with MVO. The CALLY index may provide supplementary in-hospital information on the presence and extent of MVO after pPCI.