MAGGIC in the machine: unveiling the predictive value of a conventional risk score in the ultra-long LVAD phenotype. Insights from the STELLAR register

Background The MAGGIC (Meta-Analysis Global Group in Chronic Heart Failure) risk score is widely used to predict mortality in heart failure patients. However, its applicability in ultra-long LVAD recipients—those surviving ≥5 years post-implantation—has not been assessed. As this population emerges as a distinct emerging heart failure phenotype, it is crucial to determine whether conventional risk scores retain prognostic relevance in patients receiving mechanical circulatory support for extended durations. Purpose This study aimed to evaluate the predictive performance of the MAGGIC score in patients with chronic LVAD support beyond five years and explore its role in defining the ultra-long LVAD phenotype. Methods A retrospective analysis was conducted on 136 ultra-long LVAD survivors from the STELLAR Register, a bicentric cohort including patients from two European high-volume centers. The MAGGIC score was calculated at the five-year post-implantation mark, and survival outcomes were assessed using Kaplan-Meier analysis and Cox proportional hazards models. Interaction analyses were performed to evaluate the impact of LVAD type on survival prediction. Results The median MAGGIC score at five years was 25 (IQR: 21–29). Patients with scores above the median (≥25) had significantly lower survival rates compared to those with lower scores. One-year survival was 72.1% vs. 91.5%, three-year survival was 38.4% vs. 76.2%, and five-year survival was 14.3% vs. 61.3% for high vs. low MAGGIC groups (all p<0.001). Each 1-point increase in the MAGGIC score was associated with a 9% higher mortality risk (HR 1.09, 95% CI 1.05–1.13, p<0.001). This association remained significant after adjusting for age (HR 1.07, 95% CI 1.02–1.13, p=0.008). Patients with higher MAGGIC scores exhibited a distinct ultra-long LVAD clinical presentation, characterized by older age, worse renal function (higher creatinine levels), lower ejection fraction, and greater hemodynamic compromise, despite surviving beyond five years on LVAD support. Conclusions The MAGGIC score remains a significant predictor of mortality in ultra-long LVAD survivors, suggesting that conventional heart failure risk factors retain relevance even after five years of continuous mechanical support. These findings support the hypothesis that LVAD therapy does not fully override intrinsic patient risk profiles, and that long-term LVAD recipients may represent a unique heart failure phenotype. Further refinements to existing risk models are necessary to enhance prognostic accuracy and personalized management strategies in this evolving population.