ABSTRACT
Background and Aims
β‐lactam antibiotics stand as the most profoundly impactful classes of antibacterial agents worldwide. Their widespread application has raised safety concerns over adverse events (AEs), such as blood and lymphatic system disorders (BLSDs). This study systematically evaluated and compared BLSDs associated with β‐lactam antibiotics based on real‐world data.
Methods
We reviewed the FDA Adverse Event Reporting System (FAERS) database from 2004Q1 to 2025Q1. To explore the associations between 30 FDA‐approved β‐lactam antibiotics and BLSDs, 4 disproportionality analysis algorithms were utilized, including reporting odds ratio, proportional reporting ratio, bayesian confidence propagation neural network, and multi‐item gamma poisson shrinker. Our analytical framework included subgroup analysis and sensitivity analysis, with the Weibull distribution applied to model the pattern of BLSDs over time.
Results
The total number of reported BLSDs was 7872, accounting for 6.93% of all AEs related to β‐lactam antibiotics. Notably, piperacillin/tazobactam, ceftriaxone, meropenem, cefazolin, cefepime, ceftazidime, ampicillin, oxacillin, cefotetan, and piperacillin produced strong signals of disproportionate reporting for BLSDs. Pyrexia, rash, and acute kidney injury emerged as the most frequently reported concurrent AEs. At the designated medical event level, several β‐lactam antibiotics were found to be associated with agranulocytosis, hemolytic anemia, immune thrombocytopenia, pancytopenia, and other hematological AEs. With respect to the time‐to‐onset patterns, ceftriaxone, meropenem, and ceftazidime were classified as early failure modes, whereas cefepime, cefotetan, oxacillin, and piperacillin were categorized as wearout failure modes.
Conclusions
This study revealed a notable association between β‐lactam antibiotics and BLSDs, underscoring the importance of blood tests in clinical practice when β‐lactam antibiotics are prescribed.
Keywords: β‐lactam antibiotics, blood and lymphatic system disorders, disproportionality analysis, FAERS database
1. Introduction
β‐lactam antibiotics stand as one of the most clinically influential classes of antibacterial agents globally. The discovery of the first antibiotic, Penicillin G, marked a pivotal milestone in modern antibacterial therapy [1]. Subsequently, a range of other β‐lactam antibiotics, including cephalosporins, carbapenems, β‐lactamase inhibitors, and monobactams, have been integrated into clinical practice, transforming the treatment of bacterial infections [1]. β‐lactam antibiotics exert their antimicrobial effects by irreversibly inhibiting penicillin‐binding proteins, a process that disrupts bacterial cell wall synthesis and endows them with potent activity against both gram‐positive and gram‐negative pathogens [2]. Decades of intensive clinical research and therapeutic use have solidified the excellent efficacy of β‐lactam antibiotics in addressing a wide spectrum of infectious syndromes, establishing them as the cornerstone of global antimicrobial therapy with unprecedented dominance in prescribing patterns across diverse healthcare systems [3]. Recent epidemiological data revealed that in the European Union, β‐lactam antibiotics constitute 58% of total antibiotic consumption in the community setting, with this proportion increasing to 63% in hospital environment [4]. A similar utilization trend is observed in the United States, where β‐lactams make up approximately 65% of all parenteral antibiotic prescriptions [5].
Nevertheless, the extensive clinical use of β‐lactam antibiotics has raised notable safety concerns. Clinical evidence indicates that these antimicrobial agents are associated with a spectrum of adverse events (AEs) affecting multiple organ systems, particularly the dermatological, gastrointestinal, neurological, and hematological systems [6]. The hematological toxicity profile of β‐lactam antibiotics has been systematically characterized since the initial report of piperacillin‐induced immune hemolytic anemia in the 1980s [7]. These AEs include coagulation dysfunction and bleeding, thrombocytopenia, leukopenia, granulocytopenia, and immune hemolysis, which are found to be primarily mediated through myelosuppressive effects and/or immune‐mediated mechanisms targeting hematopoietic lineages [8, 9]. Research indicates that approximately 25% of drug‐induced agranulocytosis and severe neutropenia cases are attributed to β‐lactam antibiotics; without appropriate intervention, over 60% of these patients may progress to sepsis, potentially culminating in septic shock [10]. Furthermore, β‐lactam‐induced immune hemolytic anemia can significantly impair oxygen transport, leading to tissue hypoxia and multiple organ failure, with fatal cases having been documented [11]. Notably, these drugs are also associated with thrombocytopenia and other blood and lymphatic system disorders (BLSDs) [12, 13]. Such AEs linked to β‐lactam antibiotics not only extend hospital stays and drive up healthcare costs, but also present life‐threatening risks in severe cases. Consequently, the BLSDs linked to β‐lactam antibiotics merit increased clinical attention.
Current clinical evidence on the BLSDs associated with β‐lactam antibiotics has relied primarily on clinical trials, meta‐analyzes, and case reports, with a notable lack of real‐world data support, which has introduced certain limitations. First, clinical trials are characterized by strict inclusion and exclusion criteria, well‐defined research subjects, and standardized intervention measures, which stand in contrast to the individualized medical care delivered in clinical practice [14]. This may lead to discrepancies in AE reporting between clinical research and real‐world scenarios. Second, recent meta‐analyzes focused primarily on a limited number of β‐lactam antibiotics and lacked a comprehensive comparison of their safety profiles, particularly with respect to BLSDs. Third, due to the lack of representativeness and the difficulty in clarifying the relationships between AEs and multiple contributing factors, case reports are unsuitable as a basis for clinical decision‐making [15]. In contrast, real‐world data encompass a broader patient population and more accurately reflect how drugs are used in actual clinical settings [14]. The FDA Adverse Event Reporting System (FAERS) database systematically collects spontaneous AE reports from diverse sources, including healthcare professionals, patients, and pharmaceutical manufacturers across multiple regions, thereby offering valuable insights into real‐world patterns of AE occurrence [16]. Our study aims to systematically evaluate and compare the profiles of BLSDs associated with various β‐lactam antibiotics by mining data from FAERS database.
2. Methods
2.1. Data Sources
We conducted a disproportionality analysis of BLSDs associated with β‐lactam antibiotics based on FAERS database. The drug list of β‐lactam antibiotics was obtained from the drug approval package database of FDA (https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm). 30 listed β‐lactam antibiotics, including amoxicillin, amoxicillin/clavulanic acid, ampicillin, ampicillin/sulbactam, aztreonam, benzylpenicillin, cefaclor, cefadroxil, cefalexin, cefazolin, cefdinir, cefepime, cefiderocol, cefixime, cefotetan, cefoxitin, cefpodoxime, cefprozil, ceftazidime, ceftolozane/tazobactam, ceftriaxone, cefuroxime, dicloxacillin, ertapenem, imipenem/cilastatin, meropenem, oxacillin, phenoxymethylpenicillin, piperacillin, and piperacillin/tazobactam, were used as keywords to retrieve reports in the FAERS database from 2004Q1 to 2025Q1. Drugs approved after 2004Q1 were analyzed from their approval dates.
2.2. Data Processing Procedure
The extracted AE reports were imported into SAS software for processing. Following the deduplication method recommended by the FDA, the PRIMARYID, CASEID, and FDA_DT fields were extracted from the DEMO table and sorted by CASEID, FDA_DT, and PRIMARYID. For reports sharing the same CASEID, the record with the latest FDA_DT was kept; if both CASEID and FDA_DT were identical, the record with the highest PRIMARYID was retained. Each quarterly data package included a list of reports deleted since 2019Q1. Subsequent to deduplication, reports whose CASEID matched those in the deletion list were further excluded. Finally, all AEs were categorized based on MedDRA. The system organ class (SOC) “blood and lymphatic system disorders” (code: 10005329) was selected to describe and analyze the signals of disproportionate reporting (SDRs). Concurrent AEs were identified based on the preferred terms (PTs). To determine the clinical priority of safety concerns, the public list formulated by the European Medicines Agency was employed to identify designated medical events (DMEs) [17]. These events, characterized by a confirmed drug‐related risk, could qualify as safety issues under specific conditions.
2.3. Disproportionality Analysis
This study aims to identify the signals of SDRs related to BLSDs associated with all β‐lactam antibiotics using 4 disproportionality analysis algorithms, including Reporting Odds Ratio (ROR) [18], Proportional Reporting Ratio (PRR) [19], Bayesian Confidence Propagation Neural Network (BCPNN) [20], and Multi‐item Gamma Poisson Shrinker (MGPS) [21]. All disproportionality analysis algorithms are based on a 2 × 2 contingency table (Supporting Information Table S1). In order to reduce the false positive rate, only those events that met all 4 algorithm criteria were identified as positive SDRs: the number of reports ≥ 3, the lower limit of 95% CI for ROR > 1, PRR ≥ 2 with χ2 ≥ 4, Information Component minus two Standard Deviations (IC‐2SD) > 0, and the Fifth Percentile of the Empirical Bayesian Geometric Mean Distribution (EB05) > 2 [16]. The details of the formulas are shown in Supporting Information Table S2. Subgroup analysis was carried out to examine the differences in the SDRs across the age and gender subgroups. To further verify the robustness of the findings, sensitivity analysis was performed by excluding concomitant medications. Statistical analyzes and visualizations were performed using MedDRA (version 28.0, GE, CH), Origin (version 10.1 SR1, MA, USA), SAS (version 9.4, NC, USA), and GraphPad Prism (version 8.0, CA, USA).
2.4. Onset Time Analysis
The time to onset (TTO) was defined as the time interval between the initiation date of β‐lactam antibiotics and the onset date of BLSDs. Cases in which the drug initiation date postdated the onset of BLSDs, or entries with missing dates were excluded. The Weibull distribution analysis was applied to model temporal changes in the occurrence of BLSDs.
3. Results
3.1. Descriptive Analysis
We conducted a comprehensive search of the FAERS database, encompassing the period from 2004Q1 to 2025Q1. Drugs approved after 2004Q1 were analyzed from their approval dates. Thirty β‐lactam antibiotics were identified as the primary suspected drugs linked to BLSDs across 7872 reports, accounting for 6.93% (7872/113606) of all AE reports. The incidence of BLSDs among all reported AEs varied across different β‐lactam antibiotics, with cefotetan exhibiting the highest incidence (30/67, 44.78%), followed by piperacillin (16/74, 21.62%), piperacillin/tazobactam (1554/8907, 17.45%), ceftazidime (159/1015, 15.67%), meropenem (727/4890, 14.87%), oxacillin (37/306, 12.09%), ceftriaxone (1121/10487, 10.69%), and cefepime (196/1943, 10.09%). In contrast, aztreonam demonstrated the lowest incidence (76/12128, 0.63%) among all 30 β‐lactam antibiotics (Figure 1).
Figure 1.
The occurrence of β‐lactam antibiotic related BLSDs reported in the FAERS database.
The clinical characteristics of the patients with β‐lactam antibiotic related BLSDs are described in Table 1. With respect to gender distribution, 44.8% of the BLSDs (3525/7872) involved female patients, whereas 50.0% involved male patients (3935/7872). Among patients with known ages, 2925 reports were found in elderly people (≥ 65), accounting for the highest proportion (37.2%), followed by those aged 45‐64 years (1905/7872, 24.2%). The lowest number of reports was observed in patients under 18 years of age (900/7872, 11.4%). The majority of events were reported by healthcare professionals (6317/7872, 80.2%), with the most reported cases originating from Europe (3654/7872, 46.4%). In terms of severity, the most frequently reported cases were serious BLSDs (7710/7872, 97.9%), while only 2.1% of the reports were non‐serious (162/7872). In terms of time to onset, 3532 events of BLSDs were detected within 30 days after anti‐infection treatment, representing 44.9% among all patients (3532/7872). Only 2.8% and 0.9% events occurred in 31–60 days and over 60 days after β‐lactam antibiotic administration, respectively. Furthermore, concurrent AEs were categorized using PTs. Among these, the most frequently reported PT was pyrexia, accounting for 3.51% (1064/30320). The subsequent concurrent AEs included rash (584/30320, 1.93%), acute kidney injury (492/30320, 1.62%), rash maculo‐papular (379/30320, 1.25%), and drug reaction with eosinophilia and systemic symptoms (369/30320, 1.22%), as shown in Figure 2. The involved SOCs included general disorders and administration site conditions, skin and subcutaneous tissue disorders, renal and urinary disorders, among others.
Table 1.
Demographic characteristics of patients with β‐lactam antibiotic related BLSDs in the FAERS database.
| Characteristics | Reports (N, %) |
|---|---|
| Gender | |
| Female | 3525 (44.8%) |
| Male | 3935 (50.0%) |
| Not specified | 412 (5.2%) |
| Age group (years) | |
| < 18 | 900 (11.4%) |
| 18–44 | 1268 (16.1%) |
| 45–64 | 1905 (24.2%) |
| ≥ 65 | 2925 (37.2%) |
| Not specified | 874 (11.1%) |
| Reporter | |
| Healthcare professional | 6317 (80.2%) |
| Consumer | 1223 (15.5%) |
| Not specified | 332 (4.2%) |
| Report continent | |
| Europe | 3654 (46.4%) |
| North America | 2020 (25.7%) |
| Asia | 903 (11.5%) |
| South America | 80 (1.0%) |
| Oceania | 72 (0.9%) |
| Africa | 32 (0.4%) |
| Not specified | 1111 (14.1%) |
| Severity | |
| Serious | 7710 (97.9%) |
| Non‐serious | 162 (2.1%) |
| Time to onset | |
| 0–30 days | 3532 (44.9%) |
| 31–60 days | 218 (2.8%) |
| > 60 days | 74 (0.9%) |
| Not specified | 4048 (51.4%) |
Figure 2.
The case number of concurrent AEs for β‐lactam antibiotic related BLSDs reported in the FAERS database.
3.2. Signal Detection
To evaluate the signal strength of β‐lactam antibiotic related BLSDs, we applied 4 types of disproportionality analysis algorithms, including ROR, PRR, BCPNN, and MGPS. A signal was deemed positive if its intensity significantly exceeded the background frequency across all 4 algorithms. Our study revealed that 10 β‐lactam antibiotics including piperacillin/tazobactam, ceftriaxone, meropenem, cefazolin, cefepime, ceftazidime, ampicillin, oxacillin, cefotetan, and piperacillin exhibited stronger SDRs for BLSDs compared to other β‐lactam antibiotics (Figure 3). The chemical structures for 10 antibiotics are shown in Supporting Information Figure S1. Notably, although ceftolozane/tazobactam did not meet the criterion of EB05, it is very close to the cutoff value. Therefore, the risk of BLSDs associated with ceftolozane/tazobactam should not be overlooked. By contrast, cefalexin, aztreonam, cefdinir, benzylpenicillin, phenoxymethylpenicillin, cefadroxil, cefprozil, and dicloxacillin showed weak correlations with BLSDs, as these 8 drugs failed to satisfy at least one criterion.
Figure 3.
The results of disproportionality analysis for β‐lactam antibiotic related BLSDs. The top 10 drugs satisfied all 4 disproportionality criteria; the last 8 drugs failed to satisfy at least one criterion.
3.3. Sensitivity Analysis
For patients with bacterial infections, β‐lactam antibiotics are commonly used in combination with nonsteroidal anti‐inflammatory drugs, antitussive and expectorant drugs, proton pump inhibitors, and other classes of antibiotics. After excluding reports involving concomitant medications, we found ampicillin exhibited relatively a weaker signal of BLSDs among the 10 drugs identified before sensitivity analysis (Figure 4). Except for ampicillin, patients treated with cefazolin, ceftazidime, and oxacillin were also found to be less commonly affected by BLSDs. On the contrary, piperacillin/tazobactam, ceftriaxone, meropenem, cefepime, cefotetan, and piperacillin exhibited strong SDRs of BLSDs in the sensitivity analysis.
Figure 4.
The sensitivity analysis results of the 10 β‐lactam antibiotics. Piperacillin/tazobactam, ceftriaxone, meropenem, cefepime, cefotetan, and piperacillin remained to satisfy all 4 disproportionality criteria.
3.4. Subgroup Analysis
In order to ascertain the influencing factors of BLSDs associated with β‐lactam antibiotics, we applied 4 disproportionality analysis algorisms in gender and age subgroups. Gender‐based subgroup analysis illustrated that female was less likely to be affected by BLSDs during the treatment with cefepime (N = 58, PRR = 1.66, χ2 = 15.60, EB05 = 1.28) and ampicillin (N = 36, PRR = 1.49, χ2 = 5.80, EB05 = 1.07) (Table 2). To investigate the relationship between age and BLSDs, we stratified age into 4 subgroups. We found young patients (aged < 18) exhibited weaker SDRs of BLSDs when treated with cefazolin, ampicillin, oxacillin, cefotetan, and piperacillin, as they failed to meet all 4 disproportionality analysis algorithms (Table 3). While elderly patients (aged ≥ 65) showed weaker SDRs when receiving cefepime and ampicillin treatment.
Table 2.
The subgroup analysis results of the 10 β‐lactam antibiotics based on gender.
| Drug name | Gender | N | ROR (95% CI) | PRR (χ2) | IC‐2SD | EB05 |
|---|---|---|---|---|---|---|
| Piperacillin/Tazobactam | Female | 754 | 5.19 (4.82–5.58) | 4.90 (2368.23) | 2.17 | 4.54 |
| Male | 1178 | 4.39 (4.14–4.66) | 4.11 (2822.24) | 1.95 | 3.86 | |
| Ceftriaxone | Female | 722 | 3.37 (3.13–3.63) | 3.26 (1148.08) | 1.59 | 3.03 |
| Male | 656 | 2.46 (2.27–2.66) | 2.39 (539.74) | 1.14 | 2.21 | |
| Meropenem | Female | 374 | 5.04 (4.54–5.60) | 4.77 (1128.33) | 2.08 | 4.29 |
| Male | 510 | 4.17 (3.81–4.57) | 3.92 (1132.10) | 1.83 | 3.58 | |
| Cefazolin | Female | 104 | 2.51 (2.06–3.05) | 2.46 (90.99) | 0.99 | 2.02 |
| Male | 141 | 2.41 (2.03–2.85) | 2.34 (110.74) | 0.97 | 1.98 | |
| Ceftazidime | Female | 97 | 3.74 (3.05–4.59) | 3.60 (185.02) | 1.51 | 2.94 |
| Male | 109 | 2.67 (2.20–3.24) | 2.59 (108.18) | 1.07 | 2.13 | |
| Cefepime | Female | 58 | 1.68 (1.29–2.18) | 1.66 (15.60) | 0.34 | 1.28 |
| Male | 131 | 2.36 (1.98–2.81) | 2.29 (97.48) | 0.93 | 1.92 | |
| Ampicillin | Female | 36 | 1.50 (1.08–2.08) | 1.49 (5.81) | 0.07 | 1.07 |
| Male | 83 | 3.82 (3.05–4.77) | 3.61 (160.17) | 1.48 | 2.89 | |
| Oxacillin | Female | 16 | 5.03 (3.03–8.36) | 4.76 (48.23) | 1.23 | 2.87 |
| Male | 34 | 5.12 (3.60–7.29) | 4.73 (102.04) | 1.58 | 3.32 | |
| Cefotetan | Female | 28 | 15.97 (10.59–24.07) | 13.19 (319.93) | 2.62 | 8.75 |
| Male | 4 | 5.03 (1.80–14.09) | 4.66 (11.73) | 0.05 | 1.66 | |
| Piperacillin | Female | 17 | 14.19 (8.43–23.90) | 11.97 (173.39) | 2.15 | 7.11 |
| Male | 12 | 5.21 (2.87–9.45) | 4.81 (36.92) | 1.05 | 2.65 |
Table 3.
The subgroup analysis results of the 10 β‐lactam antibiotics based on age group.
| Drug name | Age group (years) | N | ROR (95% CI) | PRR (χ2) | IC‐2SD | EB05 |
|---|---|---|---|---|---|---|
| Piperacillin/Tazobactam | < 18 | 169 | 4.39 (3.75–5.15) | 4.07 (398.84) | 1.76 | 3.46 |
| 18–44 | 406 | 6.39 (5.77–7.08) | 5.89 (1668.61) | 2.39 | 5.30 | |
| 45–64 | 552 | 4.15 (3.80–4.52) | 3.92 (1221.22) | 1.83 | 3.59 | |
| ≥ 65 | 680 | 3.35 (3.10–3.63) | 3.19 (1041.49) | 1.55 | 2.94 | |
| Ceftriaxone | < 18 | 311 | 2.62 (2.34–2.94) | 2.53 (292.17) | 1.16 | 2.24 |
| 18–44 | 173 | 2.10 (1.80–2.44) | 2.06 (96.03) | 0.81 | 1.77 | |
| 45–64 | 320 | 2.49 (2.23–2.79) | 2.43 (273.31) | 1.11 | 2.17 | |
| ≥ 65 | 493 | 2.47 (2.25–2.70) | 2.39 (407.01) | 1.12 | 2.18 | |
| Meropenem | < 18 | 117 | 3.73 (3.09–4.51) | 3.51 (214.05) | 1.50 | 2.90 |
| 18–44 | 155 | 4.87 (4.13–5.73) | 4.58 (440.91) | 1.92 | 3.89 | |
| 45–64 | 216 | 3.60 (3.14–4.13) | 3.44 (379.95) | 1.56 | 2.99 | |
| ≥ 65 | 361 | 3.86 (3.46–4.29) | 3.63 (701.62) | 1.69 | 3.25 | |
| Cefazolin | < 18 | 3 | 0.59 (0.19–1.84) | 0.60 (0.85) | −2.06 | 0.19 |
| 18–44 | 44 | 1.81 (1.34–2.45) | 1.79 (15.62) | 0.38 | 1.33 | |
| 45–64 | 73 | 2.02 (1.60–2.55) | 1.98 (36.02) | 0.62 | 1.57 | |
| ≥ 65 | 121 | 3.27 (2.72–3.94) | 3.12 (177.76) | 1.34 | 2.59 | |
| Ceftazidime | < 18 | 33 | 2.43 (1.71–3.46) | 2.35 (26.27) | 0.67 | 1.66 |
| 18–44 | 28 | 2.53 (1.74–3.70) | 2.47 (24.93) | 0.69 | 1.69 | |
| 45–64 | 56 | 3.25 (2.48–4.26) | 3.13 (82.45) | 1.20 | 2.39 | |
| ≥ 65 | 78 | 2.59 (2.06–3.26) | 2.50 (72.06) | 0.96 | 1.99 | |
| Cefepime | < 18 | 14 | 1.80 (1.05–3.07) | 1.77 (4.77) | −0.02 | 1.03 |
| 18–44 | 47 | 5.41 (4.01–7.29) | 5.05 (155.15) | 1.78 | 3.75 | |
| 45–64 | 64 | 2.28 (1.78–2.93) | 2.23 (44.13) | 0.76 | 1.73 | |
| ≥ 65 | 58 | 1.03 (0.79–1.33) | 1.03 (0.04) | −0.34 | 0.79 | |
| Ampicillin | < 18 | 13 | 1.40 (0.81–2.44) | 1.39 (1.46) | −0.35 | 0.80 |
| 18–44 | 20 | 2.08 (1.33–3.24) | 2.04 (10.77) | 0.32 | 1.31 | |
| 45–64 | 9 | 1.17 (0.61–2.27) | 1.17 (0.23) | −0.72 | 0.60 | |
| ≥ 65 | 14 | 1.13 (0.67–1.92) | 1.13 (0.21) | −0.60 | 0.66 | |
| Oxacillin | < 18 | 1 | 0.70 (0.10–5.05) | 0.70 (0.13) | −2.36 | 0.10 |
| 18–44 | 9 | 4.03 (2.05–7.91) | 3.84 (19.22) | 0.63 | 1.96 | |
| 45–64 | 30 | 8.33 (5.67–12.24) | 7.35 (167.70) | 2.05 | 5.01 | |
| ≥ 65 | 7 | 3.56 (1.65–7.69) | 3.37 (11.94) | 0.31 | 1.56 | |
| Cefotetan | < 18 | 0 | — | — | — | — |
| 18–44 | 9 | 10.03 (4.96–20.30) | 8.76 (62.91) | 1.31 | 4.33 | |
| 45–64 | 12 | 24.99 (12.61–49.53) | 17.41 (189.09) | 2.00 | 8.79 | |
| ≥ 65 | 6 | 5.01 (2.15–11.66) | 4.60 (17.30) | 0.44 | 1.98 | |
| Piperacillin | < 18 | 1 | 0.96 (0.13–6.97) | 0.96 (0.00) | −2.13 | 0.13 |
| 18–44 | 14 | 71.54 (33.09–154.68) | 33.56 (449.40) | 2.46 | 15.52 | |
| 45–64 | 3 | 3.61 (1.12–11.62) | 3.45 (5.31) | −0.42 | 1.07 | |
| ≥ 65 | 11 | 6.58 (3.49–12.40) | 5.86 (45.31) | 1.16 | 3.11 |
3.5. DME List Screening
We implemented a DME screening process to evaluate the safety signals and identified 56 positive SDRs corresponding to 10 DMEs among 30 β‐lactam antibiotics. At the DME level, agranulocytosis and hemolytic anemia were found to be the most frequently reported AEs during β‐lactam antibiotic treatment. Among these drugs, amoxicillin, piperacillin/tazobactam, ceftriaxone, amoxicillin/clavulanic acid, meropenem, cefuroxime, imipenem/cilastatin, ceftazidime, ertapenem, oxacillin, cefpodoxime, ceftolozane/tazobactam, and cefixime showed strong SDRs of agranulocytosis (Figure 5 and Supporting Information Table S3). Meanwhile, the incidence of hemolytic anemia should not be ignored in the treatment with ceftriaxone, piperacillin/tazobactam, ertapenem, cefotetan, cefalexin, cefazolin, cefepime, piperacillin, ceftazidime, cefdinir, cefpodoxime, ampicillin/sulbactam, and cefaclor. It is worth noting that cefotetan and piperacillin represented similar SDRs of hemolytic anemia and autoimmune hemolytic anemia, with extremely high values in ROR, PRR, and EB05, despite their relatively low case numbers. Furthermore, meropenem, piperacillin/tazobactam, ceftriaxone, cefalexin, and cefepime were shown to be highly associated with immune thrombocytopenia, while cefuroxime, cefalexin and, imipenem/cilastatin were found to be related to thrombotic thrombocytopenic purpura.
Figure 5.
The disproportionality analysis results of β‐lactam antibiotic related BLSDs at DME level. Deeper red indicates higher values, while darker blue indicates lower values.
3.6. TTO and Weibull Distribution Analysis
The TTO for BLSDs varied across different β‐lactam antibiotics. As shown in Table 4 and Figure 6, the median TTO values for BLSDs associated with ceftriaxone, meropenem, and ceftazidime were 6 days (IQR 2–14), 4 days (IQR 2–15), and 7 days (IQR 3–15), respectively. The Weibull shape parameters of these drugs were less than 1, indicating that the probability of developing BLSDs decreased over time, which is consistent with an early failure mode. On the contrary, the median TTO values of cefepime, cefotetan, oxacillin, and piperacillin were 16 days (IQR 4–24), 12 days (IQR 7–14), 30 days (IQR 13–30), and 8 days (IQR 6–10), respectively. Their Weibull shape parameters were all greater than 1, indicating wearout failure modes. However, given that the case numbers were only 5 for oxacillin and 2 for piperacillin, this result should be interpreted with caution. On the other hand, the median TTO values of piperacillin/tazobactam, cefazolin, and ampicillin were 9 days (IQR 3–17), 2 days (IQR 0–18), and 5 days (IQR 3–42), respectively, and Weibull distribution analysis revealed random failure modes for them. This result indicated the appropriate timing for monitoring blood tests during treatment, thereby helping to reduce the risk of iatrogenic damage caused by β‐lactam antibiotics.
Table 4.
TTO and Weibull distribution analysis of β‐lactam antibiotic related BLSDs in FAERS.
| Drug Name | N | TTO (median [IQR]) | Weibull distribution | ||
|---|---|---|---|---|---|
| Scale parameter: α (95% CI) | Shape parameter: β (95% CI) | Type | |||
| Piperacillin/Tazobactam | 1151 | 9 (3, 17) | 14.26 (13.33–15.26) | 0.96 (0.92–1.00) | Random failure |
| Ceftriaxone | 541 | 6 (2, 14) | 11.77 (10.57–13.09) | 0.90 (0.84–0.95) | Early failure |
| Meropenem | 454 | 4 (2, 15) | 13.44 (11.62–15.56) | 0.72 (0.68–0.77) | Early failure |
| Ceftazidime | 122 | 7 (3, 15) | 13.14 (9.95–17.34) | 0.69 (0.62–00.77) | Early failure |
| Cefepime | 103 | 16 (4, 24) | 19.14 (16.81–21.80) | 1.63 (1.36–1.94) | Wearout failure |
| Cefotetan | 103 | 12 (7, 14) | 12.41 (10.47–14.72) | 2.96 (1.92–4.57) | Wearout failure |
| Cefazolin | 76 | 2 (0, 18) | 12.43 (9.12–16.91) | 0.90 (0.73–1.11) | Random failure |
| Ampicillin | 11 | 5 (3, 42) | 17.42 (8.69–34.91) | 0.95 (0.59–1.53) | Random failure |
| Oxacillin | 5 | 30 (13, 30) | 24.85 (17.06–36.19) | 2.44 (1.12–5.30) | Wearout failure |
| Piperacillin | 2 | 8 (6, 10) | 8.79 (6.44–12.00) | 4.70 (1.48–14.91) | Wearout failure |
Figure 6.
Cumulative incidence of β‐lactam antibiotic related BLSDs over time.
4. Discussion
Previous studies of β‐lactam antibiotic related BLSDs were predominantly derived from clinical trials, meta‐analyzes, and case reports. However, the comprehensive comparative studies across different β‐lactam agents are still lacking. In this study, a comparative analysis was conducted to evaluate BLSDs associated with 30 β‐lactam antibiotics based on the FAERS database. The highest incidence of BLSDs within all reported AEs was for cefotetan, accounting for 44.78% (30/67), followed by piperacillin (16/74, 21.62%), and piperacillin/tazobactam (1554/8905, 17.45%). Conversely, aztreonam exhibited the lowest incidence of BLSDs among all 30 β‐lactam antibiotics, with a figure of 0.63% (76/12063). The FDA drug label has warned the risk of developing hemolytic anemia associated with cefotetan compared to other cephalosporins, with a minimum increase of threefold. A review of 85 cases indicated that a single dose of cefotetan may precipitate severe autoimmune hemolytic anemia [22]. In a retrospective study of 110 episodes of piperacillin/tazobactam therapy, the incidence of leukopenia, neutropenia, and eosinophilia were 16.3%, 10%, and 10%, respectively [23]. In contrast, the incidence of BLSDs related to aztreonam was observed to be less than 1% [24].
BLSDs represent a consistent yet infrequent category of AEs associated with β‐lactam antibiotics, mainly including decreased blood cells (leukopenia, thrombocytopenia, anemia), abnormal coagulation function, and immune‐related reactions (hemolytic anemia, eosinophilia). In this study, a total of 10 β‐lactam antibiotics showed strong SDRs of BLSDs. Further sensitivity analysis revealed significant variations in BLSDs across these β‐lactam antibiotics. Only piperacillin/tazobactam, ceftriaxone, meropenem, cefepime, cefotetan, and piperacillin demonstrated strong SDRs in sensitivity analysis, indicating a highly potential association with BLSDs. The hematological toxicities linked to β‐lactam antibiotics are thought to be mediated through a range of mechanisms, such as immune‐mediated hemolysis, perturbation of gut microbiota homeostasis, and interference with vitamin K‐dependent coagulation cascades [25, 26]. Notably, piperacillin's side chain can act as a hapten to induce drug‐dependent antibodies, thereby provoking immune hemolysis and pancytopenia [27]. This immune‐mediated toxic mechanism is consistent with the results of our study. Moreover, except for concomitant medications, the underlying concurrent AEs may exert additional influence on the risk of BLSDs. First, pyrexia, condition aggravation, dyspnea, drug ineffectiveness, pneumonia, multiple organ dysfunction syndrome, hypotension, and septic shock indicate an exacerbation of infection. This can lead to the depletion of granulocytes and platelets, which in turn aggravates the risk of BLSDs [28]. Second, allergic reactions can present with symptoms such as rash, rash maculo‐papular, drug reaction with eosinophilia and systemic symptoms, pruritus, and erythema. These reactions may lead to the destruction of blood cells, resulting in immune‐mediated hematological toxicities [29]. Third, drug accumulation caused by acute kidney injury, drug interactions, and renal failure may further increase the incidence of BLSDs. Consequently, when patients develop these disorders, vigilance for BLSDs should be maintained. In contrast, aztreonam possesses a stable unique monocyclic structure, which may reduce its capacity to bind to proteins on the surface of blood cells and form antigens, thereby rarely triggering such immune‐mediated hematological AEs [30].
Analysis of DME revealed that agranulocytosis and hemolytic anemia constituted the predominant hematological AEs. Several β‐lactam antibiotics demonstrated strong SDRs of agranulocytosis, including amoxicillin, piperacillin/tazobactam, ceftriaxone, and carbapenems (meropenem, ertapenem, imipenem/cilastatin). Recent evidence suggests that an immune‐mediated reaction serves as the primary pathogenesis for β‐lactam antibiotic related agranulocytosis [31]. Other mechanisms involve antineutrophil antibodies and leukocyte agglutination, which are induced by the accumulation of reactive metabolites [32]. Our study revealed a significant association between agranulocytosis and the use of these medications. Disproportionality analysis showed that hemolytic anemia was significantly associated with ceftriaxone, piperacillin/tazobactam, ertapenem, cefotetan, cefalexin, cefazolin, cefepime, piperacillin, ceftazidime, cefdinir, cefpodoxime, ampicillin/sulbactam, and cefaclor. Among these agents, cefotetan and piperacillin showed the strongest SDRs. A retrospective study spanning 43 years indicated that 37% of antibiotic‐induced immune hemolytic anemia were associated with cefotetan at the American Red Cross Blood Services in the Southern California Region. The proportions of other antibiotics implicated included piperacillin (15%), ceftriaxone (13%), β‐lactamase inhibitors (5%), and other cephalosporins (1%) [8]. These drugs exhibit a strong binding affinity to the red blood cell (RBC) membrane in vitro, demonstrating stability even after the washing of RBCs [33]. Cefotetan bound to RBCs can be detected for up to 98 days following the administration of the last dose [33]. Piperacillin can trigger the deposition of both IgM and IgG, which subsequently form immune complexes that deposit on the surface of RBCs. This deposition activates the complement system, ultimately inducing hemolysis [34, 35, 36]. Furthermore, immune thrombocytopenia was found to be strongly associated with meropenem, piperacillin/tazobactam, ceftriaxone, cefalexin, and cefepime, which had been found in clinical practice [12]. In patients with immune thrombocytopenia, the incidence of major bleeding was 9%, while fatal hemorrhage occurs in 0.8% of patients [37]. This necessitates particular attention. Overall, for patients undergoing treatment with the aforementioned medications, we recommend maintaining vigilant monitoring of routine blood tests to facilitate early detection and prompt intervention, thereby optimizing medication safety.
Gender disparities in medical research have long been overlooked. While it is now widely acknowledged that gender differences influence the prevalence, diagnosis, and outcomes of various disease states, the inclusion rate of male participants has remained higher than that of female participants in clinical studies over the past few decades [38]. Literature indicated that gender‐stratified reporting in pharmacovigilance studies facilitated a more nuanced understanding of sex‐specific AE patterns [39]. Studies showed that female was more prone to AEs than male, partly because their lower body weight and higher fat percentage might impact the distribution and metabolism of drugs [40]. However, this study showed that the incidence of BLSDs was higher in male (50.0%) than in female (44.8%). In patients receiving cefepime or ampicillin, female exhibited weaker SDRs of BLSDs compared with male. This stands in contrast to traditional perspectives. This may be attributed to smoking, a habit more prevalent among male, which can affect the activity of hepatic drug‐metabolizing enzymes and thereby alter drug metabolism [41]. Furthermore, biases within spontaneous reporting systems may also contribute to this outcome. To further understand the gender differences in antibiotic related BLSDs, additional clinical studies are needed to explore the impact of gender on physiology, immunity, and drug metabolism.
The analysis of AE reports revealed a distinct age‐dependent pattern in the incidence of BLSDs associated with β‐lactam antibiotics. Our findings highlight significant differences in susceptibility across age groups, with elderly patients (≥ 65 years) representing the highest proportion of reports (37.2%), followed by middle‐aged adults (45–64 years, 24.5%). Pediatric patients (< 18 years) had the lowest reporting rate (11.4%). Notably, disproportionality analysis indicated that young patients exhibited weaker SDRs of hematological toxicity when treated with cefazolin, ampicillin, oxacillin, cefotetan, and piperacillin, as these drugs did not meet all four signal detection criteria. This could imply that younger individuals may have a more robust hematopoietic reserve or different pharmacokinetic patterns of these agents. Conversely, elderly patients demonstrated higher SDRs of most β‐lactam antibiotics, possibly due to underreporting, competing comorbidities, or altered drug clearance [42]. Elderly patients often exhibit reduced renal/hepatic function, leading to prolonged drug exposure and higher toxicity risk [43]. Meanwhile, the elevated proportion of body fat combined with reduced water content alters the apparent volume of distribution of water‐soluble drugs. This physiological change results in increased concentrations of β‐lactam antibiotics, which may further exacerbate hematological toxicity [44]. Children may metabolize certain β‐lactams more efficiently, but developmental differences in immune responses could also play critical roles. Discrepancies in signal detection could arise from differences in prescribing patterns. Consequently, there are important clinical implications for elderly patients: given their increased vulnerability, close hematological monitoring is warranted when prescribing β‐lactam antibiotics, particularly those with high hematological toxicities.
This study analyzed TTO of BLSDs related to 10 β‐lactam antibiotics, using the Weibull distribution to evaluate time distributions, which provides evidence‐based guidance for optimizing antibiotic selection and monitoring strategies in clinical practice [45]. In terms of TTO, 44.9% of the BLSDs appeared within 30 days. Only 2.8% and 0.9% events occurred in 31–60 days and > 60 days, respectively. The proportion of AEs occurring within 30 days is highest and consistent with literature reports [46]. The Weibull shape parameter β values of ceftriaxone, meropenem, and ceftazidime were less than 1, indicating that the risk of BLSDs declined over time, aligning with early failure mechanism. The plasma protein binding rate of meropenem is approximately 2%, which is much lower than that of some other β‐lactam antibiotics [47]. This leads to elevated free drug concentration in circulation, potentially increasing cellular toxicity in hematopoietic tissues. In contrast, the β values of cefepime, cefotetan, oxacillin and piperacillin exceeded 1, indicating a time‐dependent increase in AEs consistent with wearout failure. Oxacillin showed the highest median TTO (30 days), suggesting its AEs developed through progressive toxicity accumulation. However, these findings should be interpreted with caution given the limited sample size. The β values of piperacillin/tazobactam, cefazolin, and ampicillin indicated a random failure pattern of AEs, where TTO remained unfixed. The median TTO of cefazolin was brief (2 days), while the broad IQR (0–18 days) suggested significant heterogeneity in individuals, encompassing both rapid and delayed AEs. Overall, clinicians should focus on monitoring and managing BLSDs during the initial treatment phase with ceftriaxone, meropenem, and ceftazidime. For wearout failure agents, implement regular monitoring for hematological toxicities during extended courses of therapy. For random failure agents, employ individualized vigilance due to their unpredictable AE profiles.
While this investigation provided valuable insights, there are several limitations that should be acknowledged. First, the FAERS database aggregates spontaneous AE reports from heterogeneous sources, including voluntary submissions by healthcare professionals (physicians, pharmacists), patients, and lawyers. Important limitations of this surveillance system include potential underreporting, incomplete case documentation, and variable data quality across reporting sources. Second, most of the reports included in this study originated from Europe, the external validity of these findings may be constrained by geographic and healthcare system heterogeneity across regions. Further research endeavors should integrate pharmacovigilance data from a wide array of geographical regions to strengthen the generalizability of safety‐related findings. Third, this study demonstrates a correlation between β‐lactam antibiotics and BLSDs. However, these observations do not suffice to establish a definitive biological causation and thus warrant further validation through clinical studies in the future.
5. Conclusions
Based on FAERS database, this study detected 10 β‐lactams with strong SDRs for BLSDs, while 8 other agents had weak BLSD associations. Agranulocytosis, hemolytic anemia, immune thrombocytopenia and pancytopenia were the most common hematological adverse events. Weibull modeling showed ceftriaxone, meropenem and ceftazidime followed an early failure pattern, whereas cefepime, cefotetan, oxacillin and piperacillin displayed a wearout failure pattern. Owing to divergent BLSD time‐to‐onset profiles, individualized routine blood monitoring is recommended when prescribing β‐lactams with prominent BLSD signals.
Author Contributions
Jing Liu: validation, investigation, writing – original draft, writing – review and editing. Min Zhang: validation, formal analysis. Zongwei Tang: software, data curation, resources. Xiaoxiao Wang: supervision. Li Mu: visualization. Wanyi Chen: funding acquisition, project administration, conceptualization. Haichi Song: methodology, conceptualization, writing – original draft, funding acquisition.
Ethics Statement
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Additional Information
Correspondence and requests for materials should be addressed to H.S. or W.C.
Transparency Statement
The corresponding authors Haichi Song affirm that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.
Supporting information
Supporting File
Acknowledgments
The authors thank the FAERS database for providing the data for this study. The authors declare that financial support was received for the research, authorship, and/or publication of this article. This study was financially supported by the Technology Innovation Project of Shapingba District (grant numbers: 2025163), National Institute of Hospital Administration (grant numbers: YLZLXZ25G004), Medical Scientific Research Project of Chongqing Municipal Health Commission (grant numbers: 2026WSJK080), Chongqing Medical Scientific Research Project (Joint project of Chongqing Health Commission and Science and Technology Bureau), (grant numbers: 2024QNXM016 and 2024ZDXM025), Chongqing Medical Youth Top Talent Project (grant number: YXQN202456), Fundamental Research Funds for the Central Universities (grant number 2021CDJYGRH‐014), and Chongqing Clinical Pharmacy Key Specialties Construction Project.
Liu J., Zhang M., Tang Z., et al., “A Disproportionality Analysis of β‐lactam Antibiotic Related Blood and Lymphatic System Disorders Based on the FDA Adverse Event Reporting System Database,” Health Science Reports 9 (2026): e73061. 10.1002/hsr2.73061.
Jing Liu, Min Zhang and Zongwei Tang contributed equally to this work and share first authorship.
Contributor Information
Wanyi Chen, Email: chenwanyi@cqu.edu.cn.
Haichi Song, Email: haichi.song.17@cqu.edu.cn.
Data Availability Statement
The data that supports the findings of this study are available in the supplementary material of this article The original contributions presented in this study are included in the article/supplementary material. Further inquiries can be directed at the corresponding authors.
References
- 1. Lima L. M., Silva B. N. M., Barbosa G., and Barreiro E. J., “β‐lactam Antibiotics: An Overview From a Medicinal Chemistry Perspective,” European Journal of Medicinal Chemistry 208 (2020): 112829. [DOI] [PubMed] [Google Scholar]
- 2. Cho H., Uehara T., and Bernhardt T. G., “Beta‐Lactam Antibiotics Induce a Lethal Malfunctioning of the Bacterial Cell Wall Synthesis Machinery,” Cell 159 (2014): 1300–1311. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Hong L. T., Downes K. J., FakhriRavari A., et al., “International Consensus Recommendations for the Use of Prolonged‐Infusion Beta‐Lactam Antibiotics: Endorsed by the American College of Clinical Pharmacy, British Society for Antimicrobial Chemotherapy, Cystic Fibrosis Foundation, European Society of Clinical Microbiology and Infectious Diseases, Infectious Diseases Society of America, Society of Critical Care Medicine, and Society of Infectious Diseases Pharmacists,” Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy 43 (2023): 740–777. [DOI] [PubMed] [Google Scholar]
- 4. Vardakas K. Z., Kalimeris G. D., Triarides N. A., and Falagas M. E., “An Update on Adverse Drug Reactions Related to β‐lactam Antibiotics,” Expert Opinion on Drug Safety 17 (2018): 499–508. [DOI] [PubMed] [Google Scholar]
- 5. Baggs J., Fridkin S. K., Pollack L. A., Srinivasan A., and Jernigan J. A., “Estimating National Trends in Inpatient Antibiotic Use Among US Hospitals From 2006 to 2012,” JAMA Internal Medicine 176 (2016): 1639–1648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Tamma P. D., Avdic E., Li D. X., Dzintars K., and Cosgrove S. E., “Association of Adverse Events With Antibiotic Use in Hospitalized Patients,” JAMA Internal Medicine 177 (2017): 1308–1315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Bressler R. B. and Huston D. P., “Piperacillin‐Induced Anemia and Leukopenia,” Southern Medical Journal 79, no. 2 (1986): 255–256. [DOI] [PubMed] [Google Scholar]
- 8. Garratty G., “Immune Hemolytic Anemia Caused by Drugs,” Expert Opinion on Drug Safety 11 (2012): 635–642. [DOI] [PubMed] [Google Scholar]
- 9. Ajit S., Prasad N. S., Brij Mohan Kumar S., et al., “Multiple Drugs Causing Severe Anemia in Infective Endocarditis,” Cardiovascular and Hematological Disorders‐Drug Targets 16 (2016): 3–5. [DOI] [PubMed] [Google Scholar]
- 10. Lagacé‐Wiens P. and Rubinstein E., “Adverse Reactions to β‐lactam Antimicrobials,” Expert Opinion on Drug Safety 11 (2012): 381–399. [DOI] [PubMed] [Google Scholar]
- 11. Wu Y., Wu Y., Guo G., Zeng J., Liu Y., and Wu Y., “Piperacillin‐Tazobactam Induced Immune Hemolytic Anemia Led to Increased Renal Impairment and Eventual Death From Multiple Organ Failure in a Patient With Hypertensive Nephropathy: Case Report and Literature Review,” BMC Nephrology 24 (2023): 173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Slaught M., Rasmussen M., Bougie D., and Aster R., “Immune Thrombocytopenia Induced by Beta‐Lactam Antibiotics: Cross‐Reactions of Responsible Antibodies With Other Beta‐Lactam Drugs,” Transfusion 61 (2021): 1600–1608. [DOI] [PubMed] [Google Scholar]
- 13. Yang J., Cai C., Pan X., et al., “A 10‐year Retrospective Study of Antibacterial‐Induced Thrombocytopenia in a Women and Children Hospital Using China Hospital Pharmacovigilance System and Visual Basic for Applications,” British Journal of Clinical Pharmacology 91 (2025): 1157–1170. [DOI] [PubMed] [Google Scholar]
- 14. Dang A., “Real‐World Evidence: A Primer,” Pharmaceutical Medicine 37 (2023): 25–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Nissen T. and Wynn R., “The Clinical Case Report: A Review of its Merits and Limitations,” BMC Research Notes 7 (2014): 264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Tang Z., Song H., Pan L., et al., “A Pharmacovigilance Study of Thromboembolism Events Associated With Vascular Endothelial Growth Factor Receptor Tyrosine Kinase Inhibitors Based on FAERS Database,” Scientific Reports 15 (2025): 25120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Gatti M., Raschi E., and De Ponti F., “Serious Adverse Events With Novel Beta‐Lactam/Beta‐Lactamase Inhibitor Combinations: A Large‐Scale Pharmacovigilance Analysis,” European Journal of Clinical Microbiology and Infectious Diseases: Official Publication of the European Society of Clinical Microbiology 40 (2021): 1169–1176. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Rothman K. J., Lanes S., and Sacks S. T., “The Reporting Odds Ratio and its Advantages over the Proportional Reporting Ratio,” Pharmacoepidemiology and Drug Safety 13 (2004): 519–523. [DOI] [PubMed] [Google Scholar]
- 19. Evans S. J. W., Waller P. C., and Davis S., “Use of Proportional Reporting Ratios (Prrs) for Signal Generation From Spontaneous Adverse Drug Reaction Reports,” Pharmacoepidemiology and Drug Safety 10, no. 6 (2001): 483–486. [DOI] [PubMed] [Google Scholar]
- 20. Bate A., Lindquist M., Edwards I. R., et al., “A Bayesian Neural Network Method for Adverse Drug Reaction Signal Generation,” European Journal of Clinical Pharmacology 54 (1998): 315–321. [DOI] [PubMed] [Google Scholar]
- 21. Dumouchel W., “Bayesian Data Mining in Large Frequency Tables, With an Application to the FDA Spontaneous Reporting System,” American Statistician 53 (1999): 177–190. [Google Scholar]
- 22. Viraraghavan R., Chakravarty A. G., and Soreth J., “Cefotetan‐Induced Haemolytic Anaemia,” Adverse Drug Reactions and Toxicological Reviews 21 (2002): 101–107. [DOI] [PubMed] [Google Scholar]
- 23. Benli A., Şimşek‐Yavuz S., Başaran S., et al., “Hematologic Adverse Effects of Prolonged Piperacillin‐Tazobactam Use in Adults,” Turkish Journal of Hematology 35 (2018): 290–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Cai L., Zhou H., Wu N., Hong L., and Lin Z., “Safety Concerns of Aztreonam: A Real‐World Disproportionality Analysis Based on FDA Adverse Event Reporting System,” Expert Opinion on Drug Safety 24 (2025): 325–344. [DOI] [PubMed] [Google Scholar]
- 25. Maquet J., Lafaurie M., Michel M., Lapeyre‐Mestre M., and Moulis G., “Drug‐Induced Autoimmune Hemolytic Anemia: Detection of New Signals in the World Pharmacovigilance Database and Risk Assessment in a Nationwide Cohort Study in France,” Blood 140 (2022): 8176–8177. [Google Scholar]
- 26. Shearer M. J., Fu X., and Booth S. L., “Vitamin K Nutrition, Metabolism, and Requirements: Current Concepts and Future Research,” Advances in Nutrition 3 (2012): 182–195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Wang L., Wang L., Wei Z., Huang C., Qin L., and Tan B., “Incidence and Clinical Significance of Drug‐Induced Antibodies and Hemolytic Anemia Associated With Beta‐Lactam Antibiotics: A Prospective Cohort Study,” Pharmacoepidemiology and Drug Safety 35 (2026): e70259. [DOI] [PubMed] [Google Scholar]
- 28. Belok S. H., Bosch N. A., Klings E. S., and Walkey A. J., “Evaluation of Leukopenia During Sepsis as a Marker of Sepsis‐Defining Organ Dysfunction,” PLoS One 16 (2021): e0252206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Inman M. D., “Bone Marrow Events in Animal Models of Allergic Inflammation and Hyperresponsiveness,” Journal of Allergy and Clinical Immunology 106 (2000): S235–S241. [DOI] [PubMed] [Google Scholar]
- 30. Kanannejad Z., Pourvali A., Esmaeilzadeh H., et al., “Diagnosis and Selection of Alternative Antibiotics in Beta‐Lactams Hypersensitivity Reactions: Current Recommendations and Challenges,” International Immunopharmacology 122 (2023): 110573. [DOI] [PubMed] [Google Scholar]
- 31. Johnston A. and Uetrecht J., “Current Understanding of the Mechanisms of Idiosyncratic Drug‐Induced Agranulocytosis,” Expert Opinion on Drug Metabolism and Toxicology 11 (2015): 243–257. [DOI] [PubMed] [Google Scholar]
- 32. Cimino C., Allos B. M., and Phillips E. J., “A Review of β‐Lactam‐Associated Neutropenia and Implications for Cross‐Reactivity,” Annals of Pharmacotherapy 55 (2021): 1037–1049. [DOI] [PubMed] [Google Scholar]
- 33. Garratty G., “Immune Hemolytic Anemia Associated With Drug Therapy,” Blood Reviews 24 (2010): 143–150. [DOI] [PubMed] [Google Scholar]
- 34. Zhao H., Chen J., and Ou G., “A Case Report of Severe Drug‐Induced Immune Hemolytic Anemia Caused by Piperacillin,” Frontiers in Immunology 15 (2024): 1478545. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Gallagher A., Roberts D., Kwok R., Terjanian T., and Mukherjee I., “A Repeat Case of Cefotetan‐Induced Hemolytic Anemia in a Surgical Patient,” Cureus 14 (2022): e30675. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Packman C. H., “The Clinical Pictures of Autoimmune Hemolytic Anemia,” Transfusion Medicine and Hemotherapy 42 (2015): 317–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Altyar A. E., Kamran S. H., Batool H., Saleem Z., and Shahid N., “Investigating Thrombocytopenia in Meropenem‐Treated Patients: A Cross‐Sectional Observational Study,” Future Journal of Pharmaceutical Sciences 11 (2025): 98. [Google Scholar]
- 38. Pinn V. W., “Sex and Gender Factors in Medical Studies: Implications for Health and Clinical Practice,” Journal of the American Medical Association 289 (2003): 397–400. [DOI] [PubMed] [Google Scholar]
- 39. Lapeyre‐Mestre M., “Considering Sex‐Specific Adverse Drug Reactions Should be a Priority in Pharmacovigilance and Pharmacoepidemiological Studies,” EClinicalMedicine 17 (2019): 100216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Zucker I. and Prendergast B. J., “Sex Differences in Pharmacokinetics Predict Adverse Drug Reactions in Women,” Biology of Sex Differences 11 (2020): 32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Shan Y., Cheung L., Zhou Y., Huang Y., and Huang R. S., “A Systematic Review on Sex Differences in Adverse Drug Reactions Related to Psychotropic, Cardiovascular, and Analgesic Medications,” Frontiers in Pharmacology 14 (2023): 1096366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Corsonello A., Pedone C., and Incalzi R., “Age‐Related Pharmacokinetic and Pharmacodynamic Changes and Related Risk of Adverse Drug Reactions,” Current Medicinal Chemistry 17 (2010): 571–584. [DOI] [PubMed] [Google Scholar]
- 43. Davies E. A. and O'Mahony M. S., “Adverse Drug Reactions in Special Populations – the Elderly,” British Journal of Clinical Pharmacology 80 (2015): 796–807. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Zazzara M. B., Palmer K., Vetrano D. L., Carfì A., and Onder G., “Adverse Drug Reactions in Older Adults: A Narrative Review of the Literature,” European Geriatric Medicine 12 (2021): 463–473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Mazhar F., Battini V., Gringeri M., et al., “The Impact of anti‐TNFα Agents on Weight‐Related Changes: New Insights From a Real‐World Pharmacovigilance Study Using the FDA Adverse Event Reporting System (Faers) Database,” Expert Opinion on Biological Therapy 21 (2021): 1281–1290. [DOI] [PubMed] [Google Scholar]
- 46. Tian X., Chen T., Lu L., et al., “Abnormal Coagulation Indicators Associated With β‐lactamantibacterial Drug: A Pharmacovigilance Study Based on the FAERS Database,” Naunyn‐Schmiedeberg's Archives of Pharmacology 399 (2025): 709–720. [DOI] [PubMed] [Google Scholar]
- 47. Tomaselli F., Maier A., Matzi V., Smolle‐Jüttner F. M., and Dittrich P., “Penetration of Meropenem into Pneumonic Human Lung Tissue as Measured by In Vivo Microdialysis,” Antimicrobial Agents and Chemotherapy 48 (2004): 2228–2232. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting File
Data Availability Statement
The data that supports the findings of this study are available in the supplementary material of this article The original contributions presented in this study are included in the article/supplementary material. Further inquiries can be directed at the corresponding authors.