Predicting citrate-related reactions in plateletpheresis donors: development and temporal validation of a prognostic score: Citrate reactions in plateletpheresis donors

Background - Citrate-related reactions are among the most frequent adverse events in plateletpheresis donors. However, practical tools for individualized risk prediction in routine donor care remain limited. We aimed to identify predictors of citrate-related reactions and to develop and validate a simple prognostic score for plateletpheresis donors. Materials and methods - We conducted a single-center observational study of plateletpheresis donors at the Transfusion Center of the Valencian Community. A derivation cohort was used to identify independent predictors and develop a point-based score from multivariable logistic regression. The score was subsequently assessed in a separate temporal validation cohort. Model performance was evaluated using discrimination, calibration, and Brier score, with logistic recalibration in the validation cohort. Results - The derivation cohort comprised 1,309 donors, of whom 305 (23.3%) experienced grade 1 citrate-related symptoms. The score included four predictors: final anticoagulant citrate dextrose (ACD) volume ≥3.2 mL/kg, final processed whole-blood volume ≥3,000 mL, female sex, and age ≤60 years. The temporal validation cohort comprised 595 donors, most of whom had also contributed a different procedure to the derivation cohort; 103 donors (17.3%) experienced citrate-related symptoms. Discrimination was moderate, with an area under the receiver operating characteristic curve (AUROC) of 0.649; the Brier score was 0.138. Observed reaction rates were 10.6%, 21.2%, and 38.5% in the low-, intermediate-, and high-risk groups, respectively. Discussion - A simple four-variable score stratified plateletpheresis donors into groups with clearly different risks of grade 1 citrate-related symptoms. The score is a descriptive risk-stratification tool and is not a validated treatment rule or pre-procedure decision aid. Prospective assessment using device-generated pre-procedure estimates and external validation across centers and apheresis platforms are warranted.