Advertisement

Hepatobiliary calculi associated with ceftriaxone treatment: An analysis of FAERS data from 2004 to 2021

Open AccessPublished:October 29, 2022DOI:https://doi.org/10.1016/j.jiac.2022.10.006

      Abstract

      Introduction

      The role of gender, age, dose and other factors in the adverse reaction process of pseudocholelithiasis caused by ceftriaxone is controversial. In this study, we further explored potential risk factors using the FAERS database.

      Methods

      The reported odds ratio (ROR) and the information component (IC) of specific candidate factors were calculated by using the ROR method and the Bayesian confidence promotion neural network (BCPNN) method respectively to detect potential risk factors in adverse events(AEs) of ceftriaxone and hepatobiliary calculi(HBC). One candidate factor will be considered as a suspicious signal, or potential risk factors if its lower limit of 95% confidence interval of ROR (ROR025) is greater than 1 and its lower limit of 95% confidence interval of IC (IC025) is greater than 0.

      Results

      A total of 764 AEs of HBC were used to this analysis to evaluate candidate risk factors: Age group, Gender, Dose. Child (1–12 years): male ROR025 = 6.64, IC025 = 2.42, female ROR025 = 6.66, IC025 = 2.40. Adolescent group (12–18 years): male ROR025 = 5.47, IC025 = 2.08; elderly (≥65 years): female ROR025 = 1.25, IC025 = 0.22.

      Conclusions

      Gender was not detected as a risk factor for HBC caused by ceftriaxone. However, Male infants, male children, female children, adolescent male, and elderly female were potential risk factors for HBC caused by ceftriaxone based on criteria ROR025 > 1 and IC025 > 0.

      Keywords

      1. Introduction

      1.1 Ceftriaxone-related pseudocholelithiasis

      Ceftriaxone (CTRX), as a third-generation cephalosporin antibiotic, is excreted through renal and biliary channels. CTRX is concentrated in the gall bladder after entering the bile, and the concentration in the bile can reach 20–150 times of the blood concentration [
      • Richards D.M.
      • et al.
      Ceftriaxone. A review of its antibacterial activity, pharmacological properties and therapeutic use.
      ]. High concentrations of CTRX in bile combine with calcium to form deposits, which can lead to cholestasis or the formation of drug stones [
      • Park H.Z.
      • et al.
      Ceftriaxone-associated gallbladder sludge. Identification of calcium-ceftriaxone salt as a major component of gallbladder precipitate.
      ].
      In 1986,Schaad et al. [
      • Schaad U.B.
      • et al.
      Transient formation of precipitations in the gallbladder associated with ceftriaxone therapy.
      ] firstly described precipitations in an adolescent patient's gallbladder associated with CTRX therapy. They proposed a new term reversible ceftriaxone-associated biliary pseudolithiasis” to describe the calculi that disappear after withdrawal of CTRX in 1988 [
      • Schaad U.B.
      • et al.
      Reversible ceftriaxone-associated biliary pseudolithiasis in children.
      ]. The Medical Dictionary for Regulatory Activities (MedDRA) also included ‘Pseudocholelithiasis’ as low-level term (LLT) for the first time in version 10.0 and was upgraded to preferred term (PT) in version 20.1.
      Cholelithiasis is rare in childhood, but CTRX-related cholelithiasis and bile mud have been widely reported in children [
      • Araz N.
      • et al.
      Pseudolithiasis due to ceftriaxone treatment for meningitis in children: report of 8 cases.
      ,
      • Bonnet J.P.
      • et al.
      Early biliary pseudolithiasis during ceftriaxone therapy for acute pyelonephritis in children: a prospective study in 34 children.
      ,
      • Herek O.
      • et al.
      Biliary pseudolithiasis in childhood: a case report.
      ,
      • Parisi G.
      • et al.
      [a case report of reversible biliary pseudolithiasis caused by ceftriaxone in childhood: does the problem deserve greater attention?].
      ,
      • Biner B.
      • et al.
      Ceftriaxone-associated biliary pseudolithiasis in children.
      ]. Araz et al. [
      • Araz N.
      • et al.
      Pseudolithiasis due to ceftriaxone treatment for meningitis in children: report of 8 cases.
      ] found that the risk of gallstones was greatly increased after the treatment dose of more than 100 mg/(kg·d) in children with high fever of meningitis in this area. Ito et al. [
      • Ito R.
      • et al.
      Experimental verification of factors influencing calcium salt formation based on a survey of the development of ceftriaxone-induced gallstone-related disorder.
      ] also considered that patients with CTRX treatment dose exceeding the normal daily dose and children under 10 years old were high-risk groups of pseudocholelithiasis. However, Heim-duthoy et al. [
      • Heim-Duthoy K.L.
      • et al.
      Apparent biliary pseudolithiasis during ceftriaxone therapy.
      ], after observing the clinical use of ceftriaxone sodium in 36 young people over 18 years old, considered that the gallstone caused by ceftriaxone sodium was independent of the dose and duration of medication. Imafuku et al. 's study [
      • Imafuku A.
      • et al.
      Risk factors of ceftriaxone-associated biliary pseudolithiasis in adults: influence of renal dysfunction.
      ] proposed that females may be an independent risk factor for adult CTRX-related biliary pseudolithiasis, but the reason for the dominant role of females in adult CTRX-related biliary pseudolithiasis remains unclear. However, gender is not considered as the influencing factor in other studies mentioned above. Therefore, in addition to the consensus that children are the high-risk group, the rest of the views on gender and dose are not consistent. Hence it is necessary to analyze the gender differences in different age groups.

      1.2 Signal detection method

      Based on spontaneous-reporting system (SRS), classical quantitative detection methods of adverse drug reaction (ADR) signal can be generally divided into two categories: frequency method and Bayesian method [
      • Rothman K.J.
      • et al.
      The reporting odds ratio and its advantages over the proportional reporting ratio.
      ,
      • Stricker B.H.
      • Tijssen J.G.
      Serum sickness-like reactions to cefaclor.
      ,
      • Bate A.
      • et al.
      A bayesian neural network method for adverse drug reaction signal generation.
      ,
      • Bate A.
      • Evans S.J.
      Quantitative signal detection using spontaneous adr reporting.
      ,
      • Kubota K.
      • et al.
      Comparison of data mining methodologies using Japanese spontaneous reports.
      ]. The basic idea is to investigate if there is a statistically significant difference between the proportion of two groups showed in Table 1.
      Table 12*2 contingency table for disproportionality.
      Reports with target AEsAll other AEs reports
      Reports with the suspected drugs/factorsab
      All other dugs/factors reportscd
      Common frequency methods include: proportional reporting ratio (PRR) method, reporting odds ratio (ROR) method, comprehensive standard method, etc. [
      • Rothman K.J.
      • et al.
      The reporting odds ratio and its advantages over the proportional reporting ratio.
      ] PRR method is the earliest and most basic method used in the early quantitative analysis of SRS, and has been used in the UK adverse reaction monitoring system. The ROR method [
      • Stricker B.H.
      • Tijssen J.G.
      Serum sickness-like reactions to cefaclor.
      ] was first proposed by the core laboratory of the Netherlands Pharmacovigilance Center, and is still widely used in the Netherlands Pharmacovigilance Center. The comprehensive standard method is currently used by the British Medicine and Healthcare Products Regulatory Agency (MHRA), so it is also called MHRA method.
      Bayesian method is based on Bayesian logic, which analyzes and describes drugs and corresponding adverse reactions through the probability distribution selected in advance. Since 1998, the World Health Organization (WHO) Uppsala Monitoring Center (UMC) in Sweden has adopted Bayesian confidence propagation neural network (BCPNN) method to detect ADR signals [
      • Bate A.
      • et al.
      A bayesian neural network method for adverse drug reaction signal generation.
      ,
      • Bate A.
      • Evans S.J.
      Quantitative signal detection using spontaneous adr reporting.
      ]. In addition, the GPS method, and the MGPS method, which is further improved and expanded on the basis of the GPS method, have been used in the early ADR signal detection work of the Food and Drug Administration (FDA) [
      • Kubota K.
      • et al.
      Comparison of data mining methodologies using Japanese spontaneous reports.
      ,
      • Almenoff J.S.
      • et al.
      Comparative performance of two quantitative safety signalling methods: implications for use in a pharmacovigilance department.
      ].
      In this study, we calculate the ROR025 and IC025 of gender, age, dose factors in the adverse events (AEs) of hepatobiliary calculi (HBC) caused by ceftriaxone using the reports in FDA Adverse Event Reporting System (FAERS) database. HBC refer to all stones occurring in the hepatobiliary system or signs caused by stones.

      2. Material and methods

      2.1 Data sources and preprocessing

      The data included in this study were from publicly available data from the FAERS Database, covering the period from the first quarter of 2004 to the fourth quarter of 2021. FAERS Quarterly Data files was downloaded from the FDA website (https://fis.fda.gov/extensions/FPD-QDE-FAERS/FPD-QDE-FAERS.html). We used “ceftriaxone” as the search term for DRUGNAME and PROD_AI entries, and preferred terms (PTs) specified in MedDRA version 23.0 as the search term for PT entries. Before retrieval, we preprocessed the data in FAERS database.
      FAERS database contains more than 60 million spontaneously reported AEs, but the database has duplicate reports and missing data. We first performed a de-duplication operation on the FAERS database according to FDA's own requirements, and deleted the data with PRIMARYID, CASEID, DRUG_SEQ as missing values. At the same time, in order to better establish the association between AEs and drugs, we deleted the data of the time of first administration after the event occurrence time. After performing the above procedures, 60,090,091 pairs of suspected drug-AE correspondences were obtained. A total of 764 reports of AE of hepatolithiasis after CTRX treatment were included in signal detection and risk factor analysis (Fig. 1).
      Fig. 1
      Fig. 1Flow chart of adverse event report screening.
      Fig. 1 shows the processing of adverse event reports in this study.

      2.2 AE signal detection method

      Use 2 × 2 contingency table (Table 1) makes statistical analysis on AE reports of suspicious drugs and other drugs, as well as suspicious risk factors and other factors. Two data mining methods, ROR and information component (IC) of BCPNN, were used to calculate the disproportionation rate to find possible risk factors. By comparing the target AEs reporting ratio of the target drug with that of all other drugs, if the ratio is greater than the set threshold, it is called disproportionality, indicating the generation of potential AE signal. The 95% confidence interval of ROR and IC is calculated according to the formula. If the lower limit of 95% confidence interval of ROR (ROR025) is greater than 1 or the lower limit of 95% confidence interval of IC (IC025) is greater than 0, the signal is considered to be significant. When both signals are positive, we think there is a suspicious signal, or risk factors. The calculation formula of ROR and its 95% confidence interval is as follows:
      ROR=a/cb/d=adbc


      95%CI=eln(ROR)±1.96(1a+1b+1c+1d)


      All analyses were performed with SAS version 9.4 and R version 4.1.3. Ethics approval and consent to participate are not applicable.

      2.3 Treatment of dose

      In this study, the daily dose was calculated by multiplying the dose and dose frequency given in the database, and the signal values were analyzed in subgroups according to the daily dose (Table 6). According to the drug instructions of CTRX, children under 12 years old need to be given drugs according to their body weight, so this study conducted subgroup analysis, extracted reports ≤12 years old, and calculated their daily dose.

      3. Results

      3.1 Descriptive analysis

      The number of record entries and signal values of each PT are shown in Table 2 (the data of PTs is from MedDRA version 23.0). Cholelithiasis, cholestasis and pseudocholelithiasis were the top three PTs, accounting for 59.16% in total. Top three ROR025 and IC025 are pseudocholelithiasis (ROR025 = 3152.00, IC025 = 5.59), bile duct stone (ROR025 = 15.85, IC025 = 3.37) and cholelithiasis (ROR025 = 13.04, IC025 = 3.61).
      Table 2Frequency distribution of CTRX-hepatobiliary calculi AEs.
      PTCountPercent (%)IC025ROR025
      Cholelithiasis27435.863.6113.04
      Cholestasis10914.272.666.79
      Pseudocholelithiasis699.035.593152.00
      Hepatitis cholestatic688.902.707.36
      Jaundice567.331.032.12
      Cholecystitis516.682.084.70
      Bile duct stone405.243.3715.85
      Hyperbilirubinaemia405.241.613.36
      Cholecystitis acute283.661.834.27
      Bile duct obstruction81.050.862.87
      Jaundice cholestatic70.92−0.360.90
      Ocular icterus60.79−0.241.06
      Jaundice neonatal30.39−0.870.77
      Yellow skin20.26--
      Cholecystitis chronic10.13--
      Gallbladder obstruction10.13--
      Hyperbilirubinaemia neonatal10.13--
      Total764100.002.797.14
      Table 2 shows the counts, proportions, IC025 and ROR025 of all PTs of CTRX-related hepatobiliary calculi.
      -The signal value cannot be calculated if the number of records is less than 3.
      Bold indicates that the signal values are meaningful: IC025 > 0, ROR025 > 1.
      The clinical characteristics of patients with HBC after CTRX treatment are shown in Table 3, Table 4. Male reports accounted for 45.29%, while female reports accounted for 47.77%. In terms of age groups, Infant group (1–12 months) accounted for 1.05%; children group (1–12 years) accounted for 1.70%; adolescent group (12–18 years) accounted for 5.89%; elderly group (>65 years) accounted for 36.4%.
      Table 3Sex and age of patients with hepatobiliary calculi of CTRX.
      hepatobiliary calculi AEsOther AEsIC025ROR025
      SexMale346 (45.29%)18383 (46.83%)−0.230.81
      Female365 (47.77%)17148 (43.68%)−0.061.02
      Unisex7 (0.92%)203 (0.52%)
      Unknown0 (0.00%)12 (0.03%)
      MISS
      Missing value in FAERS database.
      46 (6.02%)3509 (8.94%)
      AgeNeonate≤1 months8 (1.05%)308 (0.78%)−0.720.66
      Infant1–12 months13 (1.70%)592 (1.51%)0.680.65
      Child1–12 years119 (15.58%)3698 (9.42%)0.411.45
      Adolescent12–18 years45 (5.89%)925 (2.36%)0.781.91
      Adult18–65 years223 (29.19%)15898 (40.5%)−0.680.52
      Elderly≥65 years301 (39.4%)12572(32.02%)0.101.19
      MISS
      Missing value in FAERS database.
      55 (7.20%)5264(13.41%)
      Table 3 shows the clinical features of patients with hepatobiliary calculi on CTRX and any other drugs. The numbers and percentages are displayed.
      AEs, adverse events; IC025, the lower end of the 95% confidence interval of IC; ROR025, the lower limit of the 95% confidence interval of ROR.
      Bold indicates that the signal values are meaningful: IC025 > 0, ROR025 > 1.
      The age group includes the upper limit of the range and does not include the lower limit of the range (except for neonate and the elderly).
      a Missing value in FAERS database.
      Table 4Outcome, countries and regions, and dechallenge of patients with hepatobiliary calculi of CTRX.
      hepatobiliary calculi AEsOther AEsIC025ROR025
      OutcomeDeath11 (1.44%)3648 (9.29%)−3.460.08
      Life-threatening8 (1.05%)2497 (6.36%)−3.500.08
      Hospitalization345 (45.16%)14205 (36.18%)0.130.26
      Disability3 (0.39%)305 (0.78%)−2.490.16
      Congenital anomaly0 (0.00%)53 (0.14%)
      RI
      Required intervention to prevent permanent impairment/damage.
      8 (1.05%)350 (0.89%)−0.870.58
      Other serious361 (47.25%)15609 (39.76%)0.061.18
      MISS
      Missing value in FAERS database.
      28 (3.66%)2590 (6.6%)
      Countries and regionsJapan232 (30.37%)1604 (4.09%)2.478.71
      France115 (15.05%)3756 (9.57%)0.341.37
      MI (Unknown)105 (13.74%)5681 (14.47%)−0.380.76
      Spain57 (7.46%)1207 (3.07%)0.801.93
      USA39 (5.10%)10006 (25.49%)−2.750.11
      Italy29 (3.80%)2417 (6.16%)−1.230.41
      China23 (3.01%)1456 (3.71%)−0.910.53
      Portugal19 (2.49%)851 (2.17%)−0.510.73
      Cyprus8 (1.05%)2 (0.01%)1.571.27
      Greece8 (1.05%)159 (0.41%)0.028.71
      DechallengePositive243 (31.81%)5441 (13.86%)0.942.48
      Negative16 (2.09%)576 (1.47%)−0.290.87
      Does not apply95 (12.43%)5443 (13.87%)−0.330.79
      Unknown104 (13.61%)7630 (19.44%)−0.950.47
      MISS
      Missing value in FAERS database.
      306 (40.05%)20167 (51.37%)
      Table 4 shows the results of patients with hepatobiliary calculi using CTRX and any other drugs, country and dechallenge. The numbers and percentages are displayed.
      AEs, adverse events; IC025, the lower end of the 95% confidence interval of IC; ROR025, the lower limit of the 95% confidence interval of ROR.
      Bold indicates that the signal values are meaningful: IC025 > 0, ROR025 > 1.
      a Missing value in FAERS database.
      b Required intervention to prevent permanent impairment/damage.
      Result of outcome: Other serious events reports accounted for 47.25%; hospitalization reports accounted for 45.16%; life-threatening reports accounted for 1.05%; death reports accounted for 1.44%. In terms of occurrence countries and regions, reports from Japan accounted for 30.37%. Dechallenge positive reports accounted for 31.81%.

      3.2 Analysis of age and sex

      Table 3 shows the results of gender and age signal values. Male signal value result: ROR025 = 0.81, IC025 = −0.23; female signal value result: ROR025 = 1.02, IC025 = −0.06. Child signal value result: ROR025 = 1.45, IC025 = 0.41; adolescent signal value result: ROR025 = 1.91, IC025 = 0.78; elderly signal value result: ROR025 = 1.19, IC025 = 0.10.

      3.3 Gender analysis in age groups

      Table 5 shows the differences in gender distribution among different age groups. In neonates, the number of AEs reported in females was less than 3, so no signal value was reported; Male: ROR025 = 2.04, IC025 = 0.52. In the children group, Male: ROR025 = 6.64, IC025 = 2.42; Female: ROR025 = 6.66, IC025 = 2.40. In the adolescent group, males: ROR025 = 5.47, IC025 = 2.08. In the elderly group, Female: ROR025 = 1.25, IC025 = 0.22.
      Table 5Sex under age group of patients with hepatobiliary calculi of CTRX.
      hepatobiliary calculi AEsOther AEsIC025ROR025
      NeonateMale6 (0.79%)785 (0.52%)−0.730.68
      ≤1 monthsFemale1 (0.13%)562 (0.37%)--
      InfantMale8 (1.05%)389 (0.26%)0.522.04
      1–12 monthsFemale5 (0.65%)317 (0.21%)−0.141.3
      ChildMale62 (8.12%)1537 (1.01%)2.426.64
      1–12 yearsFemale52 (6.81%)1240 (0.82%)2.406.66
      AdolescentMale32 (4.19%)843 (0.55%)2.085.47
      12–18 yearsFemale10 (1.31%)2275 (1.50%)−1.120.47
      AdultMale93 (12.17%)27240 (17.92%)−0.870.51
      18–65 yearsFemale127 (16.62%)40956 (26.94%)−0.970.45
      ElderlyMale139 (18.19%)24719 (16.26%)−0.110.95
      ≥65 yearsFemale160 (20.94%)22942 (15.09%)0.221.25
      Table 5 shows the gender results by age group in patients with hepatobiliary calculi treated with CTRX or all other drugs. The numbers and percentages are displayed.
      AEs, adverse events; IC025, the lower end of the 95% confidence interval of IC; ROR025, the lower limit of the 95% confidence interval of ROR.
      -The signal value cannot be calculated if the number of records is less than 3.
      Bold indicates that the signal values are meaningful: IC025 > 0, ROR025 > 1.
      The age group includes the upper limit of the range and does not include the lower limit of the range (except for neonate and the elderly).
      Table 6Dose grouping of patients with hepatobiliary calculi of CTRX.
      hepatobiliary calculi AEsOther AEsIC025ROR025
      All age groups0–1 g/d5721932.727.84
      1–2 g/d13831273.6514.61
      2–4 g/d568163.7520.54
      >4 g/d065--
      MISS
      Missing data in FAERS database or not available through data calculation.
      51333056
      Any other drugs15200759898063−0.020.12
      ≤12 years0–20 mg/kg/d17--
      20–40 mg/kg/d232--
      40–80 mg/kg/d91591.617.76
      >80 mg/kg/d262533.1318.46
      MISS
      Missing data in FAERS database or not available through data calculation.
      1024147
      Any other drugs53091455533−0.090.10
      Table 6 shows the dose results of CTRX in patients of all ages and under 12 years of age with hepatobiliary calculi. The numbers and percentages are displayed.
      AEs, adverse events; IC025, the lower end of the 95% confidence interval of IC; ROR025, the lower limit of the 95% confidence interval of ROR.
      The dose range contains the upper limit but not the lower limit.
      -The signal value cannot be calculated if the number of records is less than 3.
      Bold indicates that the signal values are meaningful: IC025 > 0, ROR025 > 1.
      a Missing data in FAERS database or not available through data calculation.

      3.4 Analysis of onset time

      A total of 240 reports described the onset time of AEs after administration. AEs occurred on average 18.25 days after administration, and the earliest AEs occurred on the day of administration. When analyzing the onset time of different age groups, it was found that the average onset time of the elderly group was 9.99 days.

      3.5 Dosage analysis

      For all age dose analysis results: 0–1 g/d group: ROR025 = 7.84, IC025 = 2.72; 1–2 g/d group: ROR025 = 14.61, IC025 = 3.65; 2–4 g/d group: ROR025 = 20.54, IC025 = 3.75; >4 g/d group did not report signal values because the number of reports was less than 3.
      Results of dose analysis for ≤12 years: 0–20 mg/kg/d dose group and 20–40 mg/kg/d dose group did not report signal values because the number of reports was less than 3; 40–80 mg/kg/d group: ROR025 = 7.76, IC025 = 1.61; >80 mg/kg/d group: ROR025 = 18.46, IC025 = 3.13.

      3.6 Analysis of combined symptoms

      Concurrent symptom data of 619 patients from these 764 reports were also extracted in this study. A total of 253 AEs occurred in 619 patients, and 1345 cases were accumulated, with an average of 2.17 AEs per patient. 44.26% of patients had cholelithiasis, 11.15% had pseudocholelithiasis, 6.46% had bile duct stones, and 17.61% had cholestasis. Jaundice was present in 9.05% of patients, cholestatic jaundice in 1.13%, fever and vomiting in 1.94%, nausea in 1.13%, abdominal pain in 1.78%, and epigastric pain in 1.29%. At the same time, there were more severe patients with cholecystitis (8.24%), acute cholecystitis (4.52%), acute cholangitis (1.78%), and acute pancreatitis (1.62%), which may require some intervention in the treatment. Symptom study also found reports of HBC combined with urinary calculi: hydronephrosis (1.94%), postrenal renal failure (1.94%), ureteral calculi (1.13%), anuria (0.81%), urinary calculi (0.81%), nephrolithiasis (0.81%).

      4. Discussion

      4.1 HBC based on age and sex

      The results showed that children, adolescents and the elderly were considered as risk factors for pseudocholelithiasis caused by ceftriaxone. It also showed that there was no difference in gender distribution of CTRX-related HBC as a whole, but gender differences were found in different age groups. Among patients under the age of 18, male generally showed higher signal values than female, except that neonate did not participate in the discussion due to too few reports. Since the number of reports in the infant group is relatively small, although it can produce effective signal values of ROR025 and IC025, its results may also have some limitations. We can see that when an age group is subdivided by sex, it shows different results than when analyzing age alone. By comparing the results of these two analyses, it can be seen that the positive signals of the adolescent group may come from men, while women aged 12–18 years are not the risk group for such AE. In the children group, both men and women showed positive signal values, and the signal values of both groups were greater than those of the corresponding young men. This difference in age distribution may be related to the special physiological and anatomical structure of children. Children's biliary tract is thinner and longer, which makes CTRX difficult to be excreted through the biliary tract and easy to deposit in the biliary tract and gallbladder to form biliary mud or even stones [
      • Meng D.
      • et al.
      Sonographic assessment of ceftriaxone-associated biliary pseudolithiasis in Chinese children.
      ]. In the grouping of the elderly ≥65 years old, the results show that women have positive signals, and they are more likely to have the AE of CTRX-related HBC. Comparing the results of adolescents and adults, the change of this signal value with age may be related to the age difference of female hormone secretion.
      However, considering the physiological differences caused by sex among different age groups, and combining the results of this study, it may be more appropriate to consider gender and age rather than separate analysis.

      4.2 Dosage of CTRX

      Some studies considered that low dose of CTRX can avoid the occurrence of pseudolithiasis [
      • Biner B.
      • et al.
      Ceftriaxone-associated biliary pseudolithiasis in children.
      ], but the results of this study show that even 0–1 g/d dose, HBC AEs still occurred, and the signal is positive. No record of daily dose > 4 g/d was reported in the database, and the signal value > 4 g/d could not be calculated due to the missing dose. The results showed that with the increase of dose, the signal value of adverse reactions was significantly enhanced, suggesting that the occurrence of AEs in CTRX hepatobiliary system stones was dose-related.
      Meanwhile, the dose for children should be converted according to body mass in the dosing schedule, which may exceed the recommended dose when given as an adult dose. Despite missing data, a large proportion of children were exposed to a dose of more than 80 mg/kg/d. Although there is no signal value of daily dose <40 mg/kg/d for comparison, the daily dose of >40 mg/kg/d also shows a positive signal value, and the signal is stronger when it is > 80 mg/kg/d, so the higher daily dose may be the reason for the occurrence of AEs in children with CTRX related hepatolithiasis. Other previous studies in children also found that CTRX dose of more than 2 g/d or more than 100 mg/kg/d, long-term treatment duration of CTRX, rapid injection and fasting are risk factors for CTRX-related biliary pseudolithiasis [
      • Araz N.
      • et al.
      Pseudolithiasis due to ceftriaxone treatment for meningitis in children: report of 8 cases.
      ,
      • Biner B.
      • et al.
      Ceftriaxone-associated biliary pseudolithiasis in children.
      ,
      • Soysal A.
      • et al.
      Biliary precipitation during ceftriaxone therapy: frequency and risk factors.
      ,
      • Acun C.
      • et al.
      Ceftriaxone-induced biliary pseudolithiasis and urinary bladder sludge.
      ,
      • Rivera M.
      • et al.
      Ceftriaxone-associated gallbladder pseudolithiasis in a Pd patient.
      ].

      4.3 Onset time and symptoms

      The results of this study showed that the average onset time of CTRX-related HBC was 18.25 days after administration, and AEs occurred at the earliest on the day of administration, which was later than the results reported in previous studies [
      • Schaad U.B.
      • et al.
      Reversible ceftriaxone-associated biliary pseudolithiasis in children.
      ]. The discovery of HBC in the database may occur later than the actual occurrence of the calculi, since patients often do not present with biliary symptoms, such as upper abdominal pain. Rapid occurrence and rapid regression are the clinical characteristics of CTRX induced pseudolithiasis. The FAERS database did not contain the data of the end time of AEs, so the natural regression time of HBC caused by CTRX was not statistically calculated in this study.
      Many clinical reports have observed that CTRX-related HBC often have no obvious symptoms [
      • Doi Y.
      • et al.
      Usefulness of endoscopic managements in patients with ceftriaxone-induced pseudolithiasis causing biliary obstruction.
      ,
      • Choi Y.Y.
      • et al.
      Gallbladder pseudolithiasis caused by ceftriaxone in young adult.
      ,
      • Tsukagoshi S.
      • et al.
      Progressive pseudolithiasis associated with the intravenous administration of ceftriaxone in patients with central nervous system infections.
      ], so it is not easy to be detected in the early stage of medication. This study also found that only 15.19% patients with abdominal pain, nausea, vomiting and other symptoms. Therefore, it is necessary to consider whether it is necessary to use imaging methods to detect the occurrence of pseudolithiasis after stage medication in order to adjust the subsequent medication regimen. It is also necessary to consider whether the patients with asymptomatic gallbladder abnormalities found in imaging examination need drug intervention and short-term imaging review in the future.

      4.4 Country specific data

      In our research results, we described the national and regional distribution of CTRX-related HBC. However, because the database does not contain the ethnic information of patients, we cannot further obtain the physiological relationship between CTRX-related HBC and ethnic groups. However, such differences in AE reports at the country level can still reflect some phenomena.
      If we do not consider the difference in the incidence of AEs caused by ethnic differences, the examination means relying on imaging methods, the correct diagnosis of AEs themselves, and whether medical personnel can accurately report AEs in the spontaneous reporting system will also cause a large difference in the number of AE reports. CTRX-related HBC is not as obvious as the skin reaction of drug eruption or the adverse reaction of changes in such indicators as hemogram, liver and kidney function that can be monitored through laboratory examination. It needs more to be diagnosed and found by computer tomography (CT) scanning, ultrasound and other imaging examinations. Therefore, its diagnostic rate may be affected by the diagnostic and treatment habits of medical personnel and the use rate of imaging examination methods among countries.

      4.5 Prognosis and intervention of HBC

      Previous studies have reported that the reversibility of pseudolithiasis disappeared after the cessation of CTRX [
      • Schaad U.B.
      • et al.
      Reversible ceftriaxone-associated biliary pseudolithiasis in children.
      ,
      • Araz N.
      • et al.
      Pseudolithiasis due to ceftriaxone treatment for meningitis in children: report of 8 cases.
      ,
      • Bonnet J.P.
      • et al.
      Early biliary pseudolithiasis during ceftriaxone therapy for acute pyelonephritis in children: a prospective study in 34 children.
      ], and 31.81% of the drug deactivation test in this study also reported positive results, indicating that the AEs of HBC were reversed after the cessation of drugs. There have also been reports of drug or surgical intervention after CTRX associated gallstones [
      • Famularo G.
      • et al.
      Acute cholecystitis and pancreatitis in a patient with biliary sludge associated with the use of ceftriaxone: a rare but potentially severe complication.
      ,
      • Sasaki Y.
      • et al.
      Acute pancreatitis associated with the administration of ceftriaxone in an adult patient.
      ,
      • Kim S.
      • et al.
      Acute necrotizing cholecystitis: a rare complication of ceftriaxone-associated pseudolithiasis.
      ]. Due to the reversible characteristics of CTRX-related pseudolithiasis after drug withdrawal, it is suggested that in the process of clinical treatment, attention should be paid to the judgment of the symptoms of patients, to avoid unnecessary medical measures to the patients caused additional economic burden or secondary damage to the body.
      Most reports show that pseudolithiasis improves and is cured after CTRX discontinuation, but there are still case reports of patient death [
      • Doi Y.
      • et al.
      Usefulness of endoscopic managements in patients with ceftriaxone-induced pseudolithiasis causing biliary obstruction.
      ]. This study also found statistical reports of death and life-threatening outcomes in CTRX-related HBC (Table 2). Doi et al. [
      • Doi Y.
      • et al.
      Usefulness of endoscopic managements in patients with ceftriaxone-induced pseudolithiasis causing biliary obstruction.
      ] reported a case of an 82-year-old woman who was found to have small stones and sludge in the gallbladder and common bile duct on CT scan at 15 days after continuous 5 days of 5 g/d CTRX. She was diagnosed with grade II acute cholangitis and grade 1 acute pancreatitis. The patient's heart failure was exacerbated by the hydration required for acute pancreatitis. She later developed multiple organ failure and died on day 17. This suggests that the severity of comorbidities affects the intervention and outcome of patients. In the process of treatment, the patient's situation should be individualized judgment, to avoid excessive treatment of pseudocholelithiasis should also be timely given corresponding treatment measures to avoid the exacerbation of the disease. In addition, at the beginning of treatment, if the patient is judged to be a high-risk group for CTRX-related pseudocholelithiasis and the prognosis is likely to be poor, CTRX should be avoided and other antibiotics should be used for treatment.

      5. Conclusion

      On the whole, gender is not a risk factor for HBC caused by ceftriaxone (ROR025 > 1 but IC025 < 0). In different age groups, male infants (ROR025 = 2.04 > 1, IC025 = 0.52 > 0), male children (ROR025 = 6.64 > 1, IC025 = 2.42 > 0), female children (ROR025 = 6.66 > 1, IC025 = 2.40 > 0), adolescent male (ROR025 = 5.47 > 1, IC025 = 2.08 > 0), and elderly female (ROR025 = 1.25 > 1, IC025 = 0.22 > 0) were potential risk factors for HBC caused by ceftriaxone. No matter what dose of ceftriaxone, compared with other drugs, it showed positive signals in the cause of HBC, and with the increase of dose, the signal value increased. CRTX-associated HBC generally resolve spontaneously after drug withdrawal without other intervention, but for patients with severe symptoms, drug or surgical intervention may be required.

      Author contributions

      Contributor Zhaohang Li and Guanpeng Qi were responsible for data download and access. Xin Liu, Ze Xu, Juman Ma and Aijun Zhang conducted the experimental design. Xin Liu and Ze Xu wrote all the programs involved in the study. Zuojing Li, Fenfang Wei, Ling Zhong provide help for the algorithms and experimental methods involved in this paper. All authors contributed to the writing of the final manuscript.

      Declaration of competing interest

      The authors declare no conflict of interest.

      Acknowledgement

      Science and Technology Achievement Transformation Base project of Guangdong Provincial Food and Drug Administration “Guangdong Public Drug Safety Risk Management Research Base” [Project number: 2021JDB01 ].

      References

        • Richards D.M.
        • et al.
        Ceftriaxone. A review of its antibacterial activity, pharmacological properties and therapeutic use.
        Drugs. Jun 1984; 27: 469-527https://doi.org/10.2165/00003495-198427060-00001
        • Park H.Z.
        • et al.
        Ceftriaxone-associated gallbladder sludge. Identification of calcium-ceftriaxone salt as a major component of gallbladder precipitate.
        Gastroenterology. Jun 1991; 100: 1665-1670https://doi.org/10.1016/0016-5085(91)90667-a
        • Schaad U.B.
        • et al.
        Transient formation of precipitations in the gallbladder associated with ceftriaxone therapy.
        Pediatr Infect Dis. Nov-Dec 1986; 5: 708-710https://doi.org/10.1097/00006454-198611000-00026
        • Schaad U.B.
        • et al.
        Reversible ceftriaxone-associated biliary pseudolithiasis in children.
        Lancet. 1988; 2: 1411-1413https://doi.org/10.1016/s0140-6736(88)90596-x
        • Araz N.
        • et al.
        Pseudolithiasis due to ceftriaxone treatment for meningitis in children: report of 8 cases.
        Tohoku J Exp Med. Mar 2007; 211: 285-290https://doi.org/10.1620/tjem.211.285
        • Bonnet J.P.
        • et al.
        Early biliary pseudolithiasis during ceftriaxone therapy for acute pyelonephritis in children: a prospective study in 34 children.
        Eur J Pediatr Surg. Dec 2000; 10: 368-371https://doi.org/10.1055/s-2008-1072393
        • Herek O.
        • et al.
        Biliary pseudolithiasis in childhood: a case report.
        Eur J Pediatr Surg. Oct 1999; 9: 337-339https://doi.org/10.1055/s-2008-1072277
        • Parisi G.
        • et al.
        [a case report of reversible biliary pseudolithiasis caused by ceftriaxone in childhood: does the problem deserve greater attention?].
        Minerva Pediatr. Oct 1997; 49: 499-504
        • Biner B.
        • et al.
        Ceftriaxone-associated biliary pseudolithiasis in children.
        J Clin Ultrasound. Jun 2006; 34: 217-222https://doi.org/10.1002/jcu.20228
        • Ito R.
        • et al.
        Experimental verification of factors influencing calcium salt formation based on a survey of the development of ceftriaxone-induced gallstone-related disorder.
        J Infect Chemother. Dec 2019; 25: 972-978https://doi.org/10.1016/j.jiac.2019.05.020
        • Heim-Duthoy K.L.
        • et al.
        Apparent biliary pseudolithiasis during ceftriaxone therapy.
        Antimicrob Agents Chemother. Jun 1990; 34: 1146-1149https://doi.org/10.1128/aac.34.6.1146
        • Imafuku A.
        • et al.
        Risk factors of ceftriaxone-associated biliary pseudolithiasis in adults: influence of renal dysfunction.
        Clin Exp Nephrol. Jun 2018; 22: 613-619https://doi.org/10.1007/s10157-017-1493-7
        • Rothman K.J.
        • et al.
        The reporting odds ratio and its advantages over the proportional reporting ratio.
        Pharmacoepidemiol Drug Saf. Aug 2004; 13: 519-523https://doi.org/10.1002/pds.1001
        • Stricker B.H.
        • Tijssen J.G.
        Serum sickness-like reactions to cefaclor.
        J Clin Epidemiol. Oct 1992; 45: 1177-1184https://doi.org/10.1016/0895-4356(92)90158-j
        • Bate A.
        • et al.
        A bayesian neural network method for adverse drug reaction signal generation.
        Eur J Clin Pharmacol. Jun 1998; 54: 315-321https://doi.org/10.1007/s002280050466
        • Bate A.
        • Evans S.J.
        Quantitative signal detection using spontaneous adr reporting.
        Pharmacoepidemiol Drug Saf. Jun 2009; 18: 427-436https://doi.org/10.1002/pds.1742
        • Kubota K.
        • et al.
        Comparison of data mining methodologies using Japanese spontaneous reports.
        Pharmacoepidemiol Drug Saf. Jun 2004; 13: 387-394https://doi.org/10.1002/pds.964
        • Almenoff J.S.
        • et al.
        Comparative performance of two quantitative safety signalling methods: implications for use in a pharmacovigilance department.
        Drug Saf. 2006; 29: 875-887https://doi.org/10.2165/00002018-200629100-00005
        • Meng D.
        • et al.
        Sonographic assessment of ceftriaxone-associated biliary pseudolithiasis in Chinese children.
        J Int Med Res. 2010; 38: 2004-2010https://doi.org/10.1177/147323001003800614
        • Soysal A.
        • et al.
        Biliary precipitation during ceftriaxone therapy: frequency and risk factors.
        Turk J Pediatr. Oct-Dec 2007; 49: 404-407
        • Acun C.
        • et al.
        Ceftriaxone-induced biliary pseudolithiasis and urinary bladder sludge.
        Pediatr Int. Jun 2004; 46: 368-370https://doi.org/10.1111/j.1442-200x.2004.01900.x
        • Rivera M.
        • et al.
        Ceftriaxone-associated gallbladder pseudolithiasis in a Pd patient.
        Perit Dial Int. Jan-Feb 2011; 31https://doi.org/10.3747/pdi.2010.00092
        • Doi Y.
        • et al.
        Usefulness of endoscopic managements in patients with ceftriaxone-induced pseudolithiasis causing biliary obstruction.
        Case Rep Med. 2017; 3835825https://doi.org/10.1155/2017/3835825
        • Choi Y.Y.
        • et al.
        Gallbladder pseudolithiasis caused by ceftriaxone in young adult.
        J Korean Surg Soc. Dec 2011; 81: 423-426https://doi.org/10.4174/jkss.2011.81.6.423
        • Tsukagoshi S.
        • et al.
        Progressive pseudolithiasis associated with the intravenous administration of ceftriaxone in patients with central nervous system infections.
        Intern Med. Dec 1 2017; 56: 3189-3192https://doi.org/10.2169/internalmedicine.8826-17
        • Famularo G.
        • et al.
        Acute cholecystitis and pancreatitis in a patient with biliary sludge associated with the use of ceftriaxone: a rare but potentially severe complication.
        Ann Ital Med Int. Jul-Sep 1999; 14: 202-204
        • Sasaki Y.
        • et al.
        Acute pancreatitis associated with the administration of ceftriaxone in an adult patient.
        Nihon Shokakibyo Gakkai Zasshi. Apr 2009; 106: 569-575
        • Kim S.
        • et al.
        Acute necrotizing cholecystitis: a rare complication of ceftriaxone-associated pseudolithiasis.
        Pediatr Surg Int. Jun 2006; 22: 562-564https://doi.org/10.1007/s00383-006-1656-1