Admission D-dimer is associated with subsequent cardiac rupture after acute myocardial infarction: a frequency-matched case-control study

Cardiac rupture (CR) is a rare and frequently fatal mechanical complication of acute myocardial infarction (AMI). We examined whether hemostatic markers measured in the first admission blood sample were associated with CR that was subsequently diagnosed during hospitalization. This retrospective single-center case-control study arose from 8,132 AMI admissions between June 2020 and December 2025. All 71 patients with CR were included, together with 186 controls approximately frequency-matched on STEMI/NSTEMI status and otherwise randomly sampled within strata. Medical-record review confirmed that the admission echocardiogram showed no rupture in any case and that the index blood sample preceded newly developed, objectively confirmed CR in all cases. The primary model included age, sex, Killip class III-IV, STEMI status, and log2-transformed onset-to-door time; log2-transformed D-dimer was then added. All reviewer-requested analyses were post hoc and exploratory. Admission D-dimer was available in 65 cases and 183 controls; the primary complete-case analysis included 65 cases and 181 controls. As a single marker, D-dimer had the highest discrimination of the five markers examined (area under the ROC curve 0.820, 95% CI 0.755–0.877); this standalone value is distinct from the multivariable model performance reported below. After covariate adjustment, each doubling of D-dimer was associated with higher odds of subsequent CR (adjusted OR 1.60, 95% CI 1.23–2.09; P = 0.0005). Model fit improved after D-dimer was added (likelihood-ratio chi-square = 13.56; P = 0.0002). Apparent C-statistics were 0.9238 for the clinical model and 0.9362 for the augmented model (delta = 0.0124); optimism-corrected values were 0.9167 and 0.9271. The association remained directionally consistent after adjustment for albumin, flexible modeling of onset-to-door time, restriction to STEMI, additional adjustment for MLR, and Firth penalized likelihood, and persisted when ruptures recorded within the first day after arrival were excluded (39 events; OR 1.37 per doubling, 95% CI 1.02–1.83). Higher D-dimer in the first admission blood sample was independently associated with subsequently diagnosed CR in this frequency-matched case-control sample. The incremental discrimination was modest, and the results do not establish an absolute-risk model, a clinically actionable cutoff, or clinical utility. Prospective validation in an unselected AMI cohort is required.