Charlson comorbidity Index for descriptive risk stratification of long-term All-cause mortality in elderly patients with acute myocardial infarction: a retrospective cohort study

Abstract

Background Elderly patients with acute myocardial infarction (AMI) commonly present with multiple coexisting chronic conditions. The Charlson Comorbidity Index (CCI) is a well-validated tool for quantifying comorbidity burden; however, its role in descriptive long-term risk stratification among elderly patients with AMI remains insufficiently characterized. Methods We conducted a retrospective cohort study of 350 elderly patients (age ≥65 years) admitted with AMI to Sinopharm Tongmei General Hospital between January 2020 and December 2023. CCI scores were calculated from admission records, and patients were stratified into low (CCI 1–2, n = 118), moderate (CCI 3–4, n = 142), and high (CCI ≥5, n = 90) comorbidity groups. The primary endpoint was all-cause mortality during follow-up. Cox proportional hazards models were used to describe the association between CCI-defined comorbidity burden and long-term all-cause mortality. Two separate multivariable models were constructed: Model 1 incorporated the CCI total score with non-CCI covariates, and Model 2 replaced the CCI total score with selected individual comorbidity components to avoid collinearity. Results Over a median follow-up of 40.9 months (range: 1.4–57.5 months), 98 patients (28.0%) died. All-cause mortality increased stepwise across CCI groups: 16.95% in the low, 29.58% in the moderate, and 40.00% in the high comorbidity group (Log-rank P < 0.001). On univariable Cox analysis, CCI score was significantly associated with all-cause mortality (HR = 1.164, 95% CI 1.059–1.278, P = 0.002). In multivariable Model 1 (CCI + non-CCI covariates), smoking history (HR = 1.862, P = 0.003) and PCI receipt (HR = 0.497, P = 0.001) were independently associated with long-term all-cause mortality; however, CCI did not reach statistical significance after adjustment for age and other confounders (HR = 1.089, P = 0.125). Exploratory subgroup analyses suggested that patients with comorbid chronic kidney disease (CKD) or heart failure had higher observed mortality risk, but these subgroup findings were not intended to establish independent predictive ability. Conclusions In this retrospective cohort of elderly patients with AMI, higher CCI scores were associated with greater long-term all-cause mortality and adverse clinical outcomes in univariable and Kaplan–Meier analyses, supporting the use of CCI as a simple descriptive risk-stratification tool for identifying patients with high comorbidity burden. However, CCI was not independently associated with long-term mortality after multivariable adjustment; therefore, these findings should not be interpreted as confirmatory evidence that CCI independently predicts long-term mortality. Smoking history was independently associated with worse prognosis, whereas PCI receipt was associated with lower observed mortality after adjustment; the latter association should be interpreted cautiously because of potential indication bias, contraindications, treatment-selection effects, and residual confounding inherent to the observational study design. Routine CCI assessment may assist in characterizing comorbidity burden, but individualized risk assessment should integrate age, AMI severity, renal function, LVEF, PCI eligibility, treatment preferences, and other relevant clinical factors.