Cefaclor-induced hypersensitivity: Differences in the incidence of anaphylaxis relative to other 2nd and 3rd generation cephalosporins

Cefaclor, a second-generation oral cephalosporin, is the most frequently prescribed cephalosporin in Korea. Studies, however, have yet to analyze the incidence of cefaclor-associated adverse drug reactions (ADRs), including hypersensitivity (HS), according to total national usage rates. This study aimed to investigate the incidence rates and clinical features of cefaclor ADRs reported to the Korean Adverse Event Reporting System (KAERS) and Health Insurance Review and Assessment Service (HIRA) database for the most recent 5 years. Reviewing the HIRA database, which contains information on all insurance claims, including prescribed medications and patient demographics, we identified the total number of individuals who had been prescribed cefaclor and other cephalosporins including 2nd generation without cefaclor and 3rd generation antibiotics from January 2014 to December 2018. Additionally, we retrospectively analyzed all ADRs reported to the KAERS for these drugs over the same study period. Incidence rates for ADRs, HS, and anaphylaxis to cefaclor were 1.92/10,000 persons, 1.17/10,000 persons, and 0.38/10,000 persons, respectively, lower than those to other 2nd and 3rd cephalosporins. Among all ADRs, HS (60.9% vs. 43.6% vs. 44.8%, P <0.001) and anaphylaxis (19.8% vs. 4.6% vs. 4.7%, P <0.001) were more common for cefaclor than for other 2nd and 3rd cephalosporins. Females, individuals under 65 years of age, concomitant use of drugs, and serious ADRs were more strongly associated with HS to cefaclor than with HS to other 2nd and 3rd cephalosporins. In a nationwide database for the Korean population, the incidence of cefaclor-induced ADRs, particularly HS and anaphylaxis, was high. Female sex, age younger than 65 years, and concomitant use of drugs may be associated with HS to cefaclor.


Classification and assessment of adverse drug reaction data
In this study, we retrospectively reviewed all ADRs to cefaclor and other 2 nd and 3 rd cephalosporins reported to the KAERS from January 2014 to December 2018. We identified HS and anaphylaxis using WHO-ART terms. Cases of HS were defined according to WHO-ART preferred terms (PTs) as follows: A) ADR reports wherein more than one code listed in S1 Table was present or B) ADR reports wherein two or more of the codes listed in S2 Table were present. Cases of anaphylaxis were defined according to WHO-ART PTs and System Organ Classes (SOCs) as follows (S3 Table): A) ADR reports coded "anaphylaxis" among SOCs; B) ADR reports coded "skin" and "cardiovascular" or "respiratory" among SOCs; and C) ADR reports coded with two or more SOCs of "skin", "cardiovascular", "respiratory", and "gastrointestinal".
All ADR cases included in the present study were assessed as being of possible, probable, or certain cause according to the WHO-Uppsala Monitoring Center criteria by healthcare professionals in regional pharmacovigilance centers. A serious adverse reaction was defined as "one requiring or prolonging hospitalization, causing congenital anomaly, resulting in persistent or significant disability, life-threatening, or resulting in death" [4]. We reviewed information on age, sex, type of reporters, concomitant medications, and indications of drug administration in each ADR case from the KAERS database. Suspected and concomitant medications were identified using ATC classification, and indications of drug administration were identified using International Classification of Diseases 10 th Revision (ICD-10). This study was approved by the institutional review boards of both Ajou University Hospital and Dong-A University Hospital (AJRIB-MED-MDB-19-231 and DAUHIRB-19-136), respectively, and all data were fully anonymized before we accessed them.

Statistical analysis
Incidence rates for ADR, HS, and anaphylaxis to cefaclor, other 2 nd generation cephalosporins and 3 rd generation cephalosporins were calculated as for each category of ADRs in the KAERS database divided by the total number of individuals who had been prescribed these drugs in the HIRA database during the study period, and are expressed as incidence per 10,000 persons. Categorical variables are described as frequencies and proportions, and continuous variables are presented as means ± standard deviations and absolute numbers. Statistical significance was assessed using Student's t-test for continuous variables and Pearson's chi-squared test for categorical variables. P<0.05 was considered statistically significant. Statistical analyses were conducted using R program, version 3.6.3 (http://www.r-project.org).

Results
The total numbers of prescriptions of cefaclor, other 2nd cephalosporins, and 3 rd cephalosporins during the study period were 35,902,648, 15,619,584, and 17,146,861 respectively, according to the HIRA database. Annual numbers of prescriptions of cefaclor and other 2 nd and 3 rd cephalosporins are shown in Table 1. Patients who had been prescribed cefaclor or other 2 nd and 3rd cephalosporins showed similar distribution of sex and age (Table 1). In total, 6,883, 12,800, and 22,294 cases of ADRs to cefaclor, other 2 nd cephalosporins, and 3rd cephalosporins, respectively, were reported in the KAERS database during the study period. All annual reports of ADRs to cefaclor, other 2 nd cephalosporins, and 3 rd cephalosporins increased continuously, and most of the ADRs to cefaclor, other 2 nd cephalosporins, and 3 rd cephalosporins were spontaneously reported by regional pharmacovigilance centers (

Incidence rates of adverse drug reactions, hypersensitivity, and anaphylaxis to cefaclor and other 2 nd and 3 rd cephalosporins
Incidence rates of ADRs to cefaclor, other 2 nd cephalosporins, and 3 rd cephalosporins during the study period were 1.92/10,000 persons, 8.19/10,000 persons, and 13.0/10,000 persons, respectively. Incidence rates for HS (and anaphylaxis) to cefaclor, other 2 nd cephalosporins, and 3 rd cephalosporins were 1.17/10,000 persons (0.38/10,000 persons), 3.57/10,000 persons (0.38/10,000 persons), and 5.82/10,000 persons (0.61/10,000 persons), respectively ( Table 3). The majority of incidence rates for ADRs, HS, and anaphylaxis to cefaclor and 3 rd cephalosporins showed a tendency to increase gradually during the study period (Fig 1). The proportions of clinical types of ADRs were significantly different between cefaclor and other 2 nd and 3 rd cephalosporins (Fig 2): HS (60.9% vs. 43.6% vs. 44.8%) and anaphylaxis (19.8% vs. 4.6% vs. 4.7%) were more common for cefaclor than for other 2 nd and 3 rd cephalosporins with statistical significance.

Comparison of clinical characteristics between hypersensitivity/ anaphylaxis to cefaclor and other 2 nd and 3 rd cephalosporins
Females more commonly experienced HS to cefaclor than to other 2 nd and 3 rd cephalosporins. Mean age of 3 rd cephalosporins HS group was the oldest, followed by cefaclor HS, other 2 nd cephalosporins HS group, and so on. However, the proportion of individuals aged �65 years was statistically significantly lower for cases of cefaclor HS than for cases of other 2 nd cephalosporins and 3 rd cephalosporins HS (14.4% vs. 16.1% vs. 26.4%) ( Table 4). The indication for drug administration were not identified in 48.7% of cases of cefaclor HS, 21.0% of cases of other 2 nd cephalosporins HS, and 16.7% of cases of 3 rd cephalosporins HS. The three most common indications for cefaclor were diseases of the respiratory system (J00-J99, 12.5%), injury, poisoning and certain other consequences of external causes (S00-T88, 8.6%), and certain infectious and parasitic diseases (A00-B99, 5.7%). Those for other 2 nd cephalosporins use were neoplasms (ICD-code: C00-D49) (12.0%); injury, poisoning and certain other consequences of external causes (S00-T88, 8.2%); and certain infectious parasitic diseases (A00-B99, 7.7%). Those for 3 rd cephalosporins were certain infectious and parasitic diseases (A00-B99, 12.3%); diseases of the respiratory system (J00-J99, 11.1%); and diseases of the digestive system (K00-K93, 9.6%) ( Table 4).

Discussion
This is the first national survey to estimate incidence rates of cefaclor ADR, HS, and anaphylaxis. In the current study, the number of prescriptions of cefaclor was very high which are nearly same as that of other cephalosporins, and this is consistent with previous reports for Korea [19,20]. While the incidence rates of ADR, HS, and anaphylaxis to cefaclor were lower

PLOS ONE
Incidence of cefaclor-induced hypersensitivity relative to other 2 nd and 3 rd generation cephalosporins than those to other 2 nd and 3 rd cephalosporins, the proportions of HS and anaphylaxis to cefaclor among cefaclor ADRs were significantly higher than the proportions thereof to other 2 nd and 3 rd cephalosporins ADRs. Overall, the incidence rates of ADR, HS and anaphylaxis to both cefaclor and 3 rd cephalosporins increased consistently over the study period, although there is the possibility that this merely reflects increases in spontaneous reporting to the KAERS.   Several studies on the incidence of HS and/or anaphylaxis have been conducted. In a retrospective population-based analysis [7] in the United States, physician-documented cephalosporin-associated anaphylaxis occurred in five oral exposures (95% confidence interval, 1/ 1,428,571-1/96,154) among 622,456 patients exposed to 901,908 courses of treatment with oral cephalosporins and in eight parenteral exposures (95% confidence interval, 1/200,000-1/ 35,971) among 326,867 patients exposed to 487,630 courses of treatment with parenteral cephalosporins. In a large multicenter retrospective cohort study [8] in Korea, the incidence of intravenous cephalosporins-induced anaphylaxis was 6.8 per 100,000 exposures, and this is similar to our results on the incidence rate of anaphylaxis to cephalosporins (0.82 per 10,000 persons). Meanwhile, several studies on cefaclor-induced anaphylaxis have, however, reported inconsistent results for various populations [15,16,[21][22][23][24], and incidence rates of cefaclor HS and anaphylaxis are unclear. The incidence rate of cefaclor HS was reported at 1.1% among 3,000 patients taking cefaclor in a previous study by Kammer et al. [25], and this is higher than the incidence rate for cefaclor anaphylaxis (0.0038%) in our study.

Common concomitant medications
Various risk factors for drug HS have been reported previously and have largely been divided into drug-related factors and host-related factors. Drug-related risk factors include the chemical properties and molecular weights of a drug, dose, route of administration, treatment duration, and frequency of exposure. Host-related factors include age, sex, atopy, specific genetic polymorphism, and comorbidity [3,26,27]. When we compared clinical characteristics for cefaclor and other 2 nd and 3 rd cephalosporins HS groups, we noted female sex and age under 65 years to be potential risk factors for cefaclor HS. Also, we found that the proportions of cases in patients aged 40 to 50 years were much higher for HS to cefaclor than for HS to other 2 nd and 3 rd cephalosporins. These results are comparable to previously reported risk factors for drug allergy, especially IgE-mediated hypersensitivity reactions [17,28].
Cefaclor has a simple structure composed of an aminobenzyl (R1) side chain and Cl at R2, and researchers have indicated that IgE binding determinants on cefaclor encircle the whole molecule [29]. This structure may be related with the relatively high proportion of HS and anaphylaxis to cefaclor compared to other 2 nd and 3 rd cephalosporins among ADRs to them. In addition, in our study, serious ADRs were significantly more common among cases of HS to cefaclor than to other 2 nd and 3 rd cephalosporins. Although we are unable to state that serious ADRs are a cefaclor-specific problems, we do suggest that cefaclor HS could potentially pose a significant socioeconomic burden and public health problem and that further research is needed to determine whether these severe reactions are associated with structural features of cefaclor.
The use of concomitant medications, including analgesic drugs, antihistamine, gastrointestinal agents, NSAIDs, and respiratory agents, were more frequently prescribed in cases of cefaclor HS than in cases of other 2 nd and 3 rd cephalosporins HS. These medications were generally prescribed to patients with respiratory disease, in particular upper respiratory infections, and diseases of the respiratory system were the most common indications for cefaclor administration in this study. This difference could affect the difference in concomitant medications between cefaclor HS and other 2 nd and 3 rd cephalosporins HS.
There are several limitations to this study. First, this study was conducted retrospectively for data reported to a national spontaneously reporting system, and there were some ADRs that were not reported in this study. In addition, prescriptions of cefaclor and other 2 nd and 3 rd cephalosporins from the HIRA database were ascertained from de-identified data, and we were unable to confirm the exact number of patients exposed to each drug. Therefore, the incidence rates of ADRs may not be exact. However, we compared the incidence rates of cefaclor ADRs with those of other 2 nd and 3 rd cephalosporins ADRs, which were estimated using same method. Although the exact incidences of ADRs were not obtained in this study, our results can emphasize relatively higher proportions of HS and anaphylaxis to cefaclor than to other 2 nd and 3 rd cephalosporins among ADRs to them using nationwide data. Second, the KAERS database contained almost all spontaneous ADR reports from mainly 27 pharmacovigilance centers, other general hospitals, pharmaceutical companies, and pharmacies. Therefore, assessment of ADRs, including causality and seriousness could differ depending on the reporter. To offset this difference as much as possible, healthcare specialists and/or professional pharmacists at regional pharmacovigilance centers conduct a final review before inclusion in the database. However, in this study, the proportions of urticaria in the clinical manifestations of anaphylaxis to cefaclor and other 2 nd and 3 rd cephalosporins were significantly lower than what is commonly known [30]. This may be because of missing information in the ADR reports. In addition, information on concomitant medications and indications for drug administration were not exact. Systematic approaches to improve the quality of the reports of ADRs are necessary. Finally, we could not evaluate potential risk factors for ADRs, including prior exposure history or exposed intensity, because the individual patients were not identified.
Notwithstanding, this is the first study to estimate incidence rates of cefaclor ADR, HS, and anaphylaxis from a nationwide pharmacovigilance database of the Korean population. Compared to other 2 nd cephalosporins and 3 rd cephalosporins, cefaclor induced relatively high proportions of HS and anaphylaxis among ADRs in this study. Female sex and age under 65 years and concomitant administration of drugs were found to be potential risk factors for cefaclor HS. Physicians should be cautious about prescribing cefaclor to patients who may be fatal to anaphylaxis or are at a relatively high risk of developing hypersensitivity reaction. Further studies to obtain the exact incidence rates of cefaclor ADRs and to confirm risk factors thereof are needed.
Supporting information S1