Reported mitral annular calcification grade and in-hospital mortality: a retrospective linked echocardiographic cohort study

Abstract Background The short-term prognostic meaning of routinely reported mitral annular calcification (MAC) grade during hospitalization remains uncertain, especially when unreported MAC cannot be equated with imaging-confirmed absence. Methods This retrospective linked echocardiographic cohort study analyzed MIMIC-IV-ECHO linked to MIMIC-IV hospital and ICU data. Admissions with inpatient transthoracic echocardiography and structured MAC severity were included. The primary analysis compared moderate and severe reported MAC with mild MAC. Logistic regression used subject-level cluster-robust standard errors and adjusted for demographics, admission factors, prior comorbidities, index diagnoses, ICU stay, ventilation, vasopressor exposure, first-day SOFA, and procedure markers. Results The cohort included 83436 admissions from 59586 subjects, with 5669 in-hospital deaths. Among admissions with reported MAC, mortality was 7.0% with mild MAC, 8.7% with moderate MAC, and 10.7% with severe MAC. The reported-MAC-only analysis included 27964 admissions and 2115 deaths. In the clinical and acute-support adjusted model, moderate MAC (odds ratio [OR] 1.28, 95% confidence interval [CI] 1.13–1.45; p < 0.001) and severe MAC (OR 1.49, 95% CI 1.20–1.85; p < 0.001) were associated with higher mortality than mild MAC. In an echocardiographic complete-case sensitivity model, the moderate-MAC association persisted (OR 1.19, 95% CI 1.02–1.39), whereas the severe-MAC estimate was less precise (OR 1.29, 95% CI 0.97–1.71). Conclusions Routinely reported moderate and severe MAC grades were associated with higher in-hospital mortality than mild MAC. Reported MAC grade should be interpreted as a marker of short-term inpatient risk context rather than causal evidence or confirmed absence-versus-presence stratification. Trial registration Not applicable.