AimsWe aimed to investigate antibacterial‐induced thrombocytopenia using the China Hospital Pharmacovigilance System (CHPS) in conjunction with Visual Basic for Applications (VBA).MethodsBetween September 2011 and December 2022, a 2‐phase workflow was employed to identify antibacterial‐induced thrombocytopenia, including preliminary screening in phase (I) conducted by CHPS algorithms and causality assessment by trained pharmacists in phase (II) using VBA. The incidence of thrombocytopenia in each antibacterial was calculated, and comparisons were performed between paediatric and adult patients.ResultsCHPS algorithms identified 4080 cases from 485 238 admissions (including 223 735 admissions receiving at least 1 antibacterial treatment). After ruling out cases with chemotherapy and abnormal platelet count at admission, 3832 cases were available. Using VBA, pharmacists identified 1039 cases (1246 antibacterial treatments, 28 agents) as potential thrombocytopenia instances (κ = 0.89), with an incidence of 0.46%. All antibacterial treatments correlated temporally with thrombocytopenia. Carbapenems (meropenem 1.77%), glycopeptides (vancomycin 1.55%) and lincosamides (clindamycin 0.44%) were prominent causal groups. The highest incidences of thrombocytopenia in the cephalosporins and penicillins groups were ceftazidime (2.04%) and piperacillin/tazobactam (1.24%), respectively. Among all antibacterial treatments, clindamycin showed the shortest time to onset (TTO), and erythromycin showed the longest TTO. Paediatric patients exhibited a longer TTO (61 vs. 29 h), extended time to nadir (83 vs. 37 h), lower platelet nadir count values (110 vs. 92 × 109/L), and a higher severe case proportion (12.37 vs. 3.86%) when compared with adults.ConclusionDifferent antibacterial agents exhibit varying incidences of thrombocytopenia, with notable disparities between adults and children in the characteristics of thrombocytopenia.
