Early Haemoglobin Response after Single-Unit Red Blood Cell Transfusion Predicts Repeat Transfusion in Hospitalised Adults: A Prospective Cohort Study and Clinical Prediction Model.

BACKGROUND:Restrictive transfusion strategies and patient blood management programs increasingly recommend administration of single-unit red blood cell (RBC) transfusions followed by reassessment. However, clinicians often lack objective indicators to determine whether additional transfusion may subsequently be administered after the initial unit. Early haemoglobin response may provide clinically useful information regarding post-transfusion haematologic response and the likelihood of repeat transfusion. METHODS:We conducted a prospective observational cohort study including 518 hospitalized adults who received single-unit RBC transfusion. Haemoglobin concentrations for the primary analysis were measured at baseline and 24 h following transfusion. The primary predictor variable was the 24-h haemoglobin increment (ΔHb24). The primary outcome was repeat RBC transfusion within 48 h. A multivariable logistic regression model incorporating haemoglobin response and clinical variables was developed. Model performance was evaluated using receiver operating characteristic analysis, calibration assessment, and decision curve analysis. RESULTS:Among the 518 patients included in the study, 42 (8.1%) received repeat transfusion within 48 h. Patients who received additional transfusion demonstrated significantly smaller haemoglobin increments compared with those who did not. In multivariable analysis, ΔHb24 emerged as the strongest independent predictor of repeat transfusion, with increasing haemoglobin increments associated with lower risk. The prediction model demonstrated moderate discrimination (AUC 0.74) and acceptable calibration. Risk stratification showed that patients with ΔHb24 <1 g/dL had substantially higher probability of repeat transfusion, whereas those with ΔHb24 >1.5 g/dL had low risk. Decision curve analysis demonstrated positive clinical net benefit across clinically relevant threshold probabilities. CONCLUSION:Early haemoglobin response may provide clinically useful prognostic information regarding repeat transfusion risk following single-unit RBC transfusion. These findings support structured reassessment following single-unit transfusion and demonstrate that early haemoglobin response may assist in identifying patients more likely to receive repeat transfusion.