Anthony Altobelli, Laura Parker, Deepa Iyer, Sai Somisetty, Maya Guglin
Background/Objectives: Accurate size-matching between donor and recipient is central to adult heart transplantation, yet the optimal operational metric remains unsettled. We compared the predicted heart mass ratio (PHMr) with a ratio pairing measured donor left ventricular mass and predicted recipient left ventricular mass (mLVM:pLVM) to determine how each classifies the size acceptability of donor offers. Methods: In a retrospective single-center cohort of 594 adult donor offers evaluated for 22 recipients during 2023, offers were classified as acceptable under bilateral windows (PHMr 0.86 to 1.14, mLVM:pLVM 0.80 to 1.20) and under lower-bound-only thresholds. Recipient-clustered models with cluster-robust standard errors were the primary analysis, with a generalized estimating equation and a linear mixed model as sensitivity analyses. Prespecified subgroups were defined by sex and body mass index. Results: Under bilateral windows, PHMr classified more offers as acceptable than mLVM:pLVM (46.3% versus 40.1%), but the overall difference was not significant in the primary analysis (cluster-robust risk difference 6.2 percentage points, 95% CI -0.5 to 13.0, p = 0.068) and reached significance only in sensitivity analyses. Agreement was low (Cohen's κ 0.12 to 0.24). The between-method difference varied by donor sex, confirmed by interaction testing: PHMr classified more male donor offers as acceptable, whereas mLVM:pLVM classified more female donor offers as acceptable. Under lower-bound-only thresholds, significant differences were confined to donor sex. Conclusions: Metric and threshold choice changed which offers were classified as acceptable. These differences reflect operational discordance rather than clinical superiority and warrant validation linked to outcomes.