Adverse reactions associated with oral and parenteral use of cephalosporins: A retrospective population-based analysis Eric Macy, MS, MD,a and Richard Contreras, MSb

San Diego and Pasadena, Calif

Background: Few studies have provided population-based, route-specific data on allergy to cephalosporin or incidence of serious adverse drug reactions (ADRs). Objective: We investigated the incidence of new reports of cephalosporin-associated ‘‘allergy’’ and serious ADRs. Methods: We identified all members of the Kaiser Permanente Southern California health plan given cephalosporins (from January 1, 2010, through December 31, 2012), all new reports of cephalosporin-associated allergy, and all serious ADRs. Results: There were 622,456 health plan members exposed to 901,908 courses of oral cephalosporins and 326,867 members exposed to 487,630 courses of parenteral cephalosporins over the 3-year study period. New reports of allergy to cephalosporin were more frequent among women (0.56%; 95% CI, 0.54% to 0.57%) than among men (0.43%; 95% CI, 0.41% to 0.44%) per course (P < .0001). The most frequent serious cephalosporinassociated ADRs were Clostridium difficile infection within 90 days (0.91%), nephropathy within 30 days (0.15%), and all-cause death within 1 day (0.10%). None correlated with history of drug allergy. Physician-documented cephalosporinassociated anaphylaxis occurred with 5 oral exposures (95% CI, 1/1,428,571-1/96,154) and 8 parenteral exposures (95% CI, 1/200,000-1/35,971) (P 5 .0761). There were 3 documented cephalosporin-associated serious cutaneous adverse reactions (95% CI, 0-1 in 217,291). All were associated with the use of another antibiotic at the same time as cephalosporin. Conclusions: Cephalosporins are widely and safely used, even in individuals with a history of penicillin allergy. Physiciandocumented cephalosporin-associated anaphylaxis and serious cutaneous adverse reactions are rare compared with C difficile infection within 90 days, nephropathy within 30 days, and all-cause death within 1 day. (J Allergy Clin Immunol 2014;nnn:nnn-nnn.) Key words: Adverse drug reaction, allergy, cephalosporin, Clostridium difficile, death, hemolytic anemia, hospitalized, oral, outpatient, nephropathy, parenteral, serious cutaneous adverse reaction, Stevens-Johnson syndrome

From athe Department of Allergy, Southern California Permanente Medical Group, San Diego Medical Center, San Diego; and bthe Department of Research and Evaluation, Kaiser Permanente Health Care Program, Pasadena. The Kaiser Permanente Health Care Program funded this project. Disclosure of potential conflict of interest: The Southern California Permanente Medical Group and the Kaiser Permanente Health Care Program employ the authors. E. Macy has received a research grant from ALK Abello, Inc, is a member of a clinical trial safety and monitoring committee for BioMarin Pharmaceutical, Inc, and has received participation fees from BioMarin. R. Contreras declares that he has no relevant conflicts of interest. Received for publication March 26, 2014; revised July 7, 2014; accepted for publication July 9, 2014. Corresponding author: Eric Macy, MS, MD, Department of Allergy, Kaiser Permanente, 7060 Clairemont Mesa Blvd, San Diego, CA 92111. E-mail: [email protected]. 0091-6749/$36.00 Ó 2014 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaci.2014.07.062

Abbreviations used ADR: Adverse drug reaction EHR: Electronic health record ICD-9: International Classification of Diseases, Ninth Revision RR: Relative risk SCAR: Serious cutaneous adverse reaction SJS: Stevens-Johnson syndrome

Cephalosporins are the most commonly used antibiotic in patients hospitalized in the Kaiser Permanente Health Care Program.1 Cephalosporins are widely used to prevent infections in patients undergoing surgery and in outpatients with genitourinary tract, upper respiratory tract, and skin infections. However, there are few infections for which cephalosporins are the antibiotic of choice. The use, and specifically overuse, of cephalosporins has been associated with adverse drug reactions (ADRs),2 ranging from rashes and diarrhea to anaphylaxis, serious cutaneous adverse reactions (SCARs), hemolytic anemia, nephropathy, Clostridium difficile infection, and death. The rate of anaphylaxis associated with parenteral cephalosporin exposure has been assumed to be greater than that associated with oral use, but this has never been comprehensively demonstrated in a large, diverse population.3 The relative frequencies of serious cephalosporin-associated ADRs are poorly understood. Many physicians still fear cephalosporin-associated anaphylaxis in individuals with a history of penicillin ‘‘allergy,’’ rather than serious and frequent morbidity from C difficile infection. Cephalosporins are currently widely, safely, and appropriately used in individuals with histories of penicillin allergy, despite concerns about immune cross-reactivity with both drugs.3 One reason cephalosporins are safely used could be that only a small minority of individuals with histories of penicillin allergy show responses to penicillin in skin tests or with oral challenges.4 There are doubts about a potential small increase in the rate of serious allergic reactions to cephalosporins among individuals with proven IgE-mediated allergy to penicillin.5 There have been a number of studies, involving as many as tens of thousands of individuals, of the incidence of antibioticassociated ADRs. These reported new cases of allergy and anaphylaxis, but most of these studies focused on penicillins and sulfonamides—classes of antibiotics with the highest prevalence of allergy.6-8 In 1983, Alanis and Weinstein9 reviewed adverse reactions associated with the use of oral penicillins and cephalosporins in the preceding 30 years but provided no data on parenteral cephalosporin-associated ADRs.9 We reported new cases of antibiotic allergy in 411,543 individuals in 2007.10 We noted that 0.60% of male outpatients (95% CI, 0.47% to 0.72%) and 1.08% of female outpatients 1

2 MACY AND CONTRERAS

(95% CI, 0.96% to 1.21%) who received cephalosporin (almost all oral) reported a new allergy to the drug within 1 year after each course of use. In a study of 2,375,424 patients seen during 2009, we associated sex, age, medication use, and history of drug allergy with future reports of antibiotic allergy.11 We now aim to provide a descriptive report of the real-world frequency of new reports of cephalosporin-associated ‘‘allergies’’ or serious ADRs, including physician-documented anaphylaxis and SCARs, along with reports of hemolytic anemia and nephropathy within 30 days, C difficile infection within 90 days, and all-cause death within 1 day.

METHODS The Kaiser Permanente Southern California Institutional Review Board reviewed and approved this project. All individuals who received health care from Kaiser Permanente in Southern California between January 1, 2010, and December 31, 2012, were potential cohort members. Subsets of these individuals have been previously reported on, in part.1 Individuals of unknown sex were not included (n 5 3133, 0.078%). There is 1 electronic health record (EHR) system used by Kaiser Permanente in Southern California, called Health Connect (Epic Systems Corporation, Verona, Wis). Health Connect links all health care provided in the outpatient and inpatient settings with pharmacies. Clinical visits were defined as an outpatient visit and included visits to medical offices, urgent care centers, infusion centers, or emergency rooms; these also included phone calls or electronic (e)-visits (via direct Health Connect electronic messages) or admission to 1 of 12 Kaiser Permanente Southern California hospitals. A small fraction of the cohort may have received health care outside the network, but if medications used or any ADRs experienced were entered into Health Connect via an e-visit or future in-person visit, the event was captured for this study. All cephalosporin antibiotics used by the cohort during the study interval were identified. All cephalosporin use was linked to an outpatient visit, e-visit, or hospitalization. Drug class allergies were determined as previously described.11 The maximum number of drug class allergies possible was set as 24. A course of a cephalosporin was defined as any oral or parenteral exposure to a specific cephalosporin by 1 systemic route. Parenteral exposures included intravenous, intramuscular, subcutaneous, intradermal, intraosseous, intraperitoneal, and intravesicular injections or infusions. Any exposure to the same cephalosporin by the same route within 36 hours was considered to be the same course. We collected data on active drug allergies noted in the EHR for each individual on the day before the use of each cephalosporin class antibiotic. Any serious, or clinically significant, reproducible IgE-, IgG-, or T-cell2mediated cephalosporin-associated ADRs, occurring because of pre-existing immunologic reactivity to cephalosporins or cross-reactivity with penicillins, would be expected to be manifest within 5 days of the exposure and could therefore be attributed to that course. To allow for potential new sensitization, we identified all new cephalosporin class antibiotic allergy entries made within 30 days of each cephalosporin use. These methods differ from those we published previously and should more accurately link drug allergy entries with specific cephalosporin courses.10 The following algorithms were used to identify cephalosporin-associated serious ADRs, including physician-documented anaphylaxis and SCARs, hemolytic anemia, nephropathy, C difficile infection, and death. We identified all individuals for whom a diagnosis of anaphylaxis (International Classification of Diseases, Ninth Revision [ICD-9] 995.0) was coded within 1 day of starting a course of cephalosporin, along with all intramuscular epinephrine administrations made within 1 day of starting the cephalosporin course. It was not possible to reliably identify intravenous use of epinephrine in the operating room. An EHR review was then performed to verify that anaphylaxis occurred and met the current working definition.12 The EHR review was also used to rule out the possibility that an alternate medication caused the anaphylaxis episode. Physician-documented SCARs were identified by using the following ICD-9 codes, entered in the EHR up to 30

J ALLERGY CLIN IMMUNOL nnn 2014

days after each course of cephalosporin began: erythema multiforme major (ICD-9 695.14), Stevens-Johnson syndrome (SJS) (ICD-9 695.13), toxic epidermal necrolysis (ICD-9 695.15), and drug reaction with eosinophilia and systemic symptoms (ICD-9 995.27). Drug reaction with eosinophilia and systemic symptoms cases were required to have more than 10% eosinophilia, with 500 or more eosinophils/mL, and increased levels of liver enzymes within 30 days of the implicated starting does of cephalosporin. An EHR review was then performed to verify that the coded SCAR diagnosis was correct and that biopsy analyses supported the diagnosis, as well as to rule out the use of another medication at the same time as the implicated cephalosporin. Potential cases of cephalosporin-associated hemolytic anemia were identified on the basis of a combination of hemoglobin levels newly below 8 gm/dL, associated with lactate dehydrogenase levels of more than 2000 U/L, within 30 days of starting the implicated cephalosporin course, along with the lack of a previous diagnosis of anemia, sepsis, or renal failure. EHRs were reviewed to confirm cases and determine clinical outcomes. Cephalosporin-associated nephropathy was identified in individuals who had new increases of more than 3.0 mg/dL in serum creatinine levels, within 30 days of starting the implicated cephalosporin course, and no history of increased serum creatinine levels (>1.3 mg/dL) or chronic kidney disease (ICD-9 codes 580 through 589 or 250.4) in the preceding 90 days. EHRs were reviewed to determine clinical outcomes. We identified all new-onset cases of clinically significant C difficile infection (verified by the first use of ICD-9 code 8.45 as a diagnosis) or a new positive result from a laboratory test for C difficile, toxin, or culture, within 90 days of using a cephalosporin. Finally, we identified all individuals who died from all causes within 1 day of starting a course of cephalosporin. All statistical analyses were conducted with SAS EG 4.3. Percentages and odds ratios were calculated on categorical outcomes. Means and SDs were calculated on continuous variables. Statistical significance was assessed on the basis of categorical variables using chi-square tests and t tests for continuous variables. All P values of less than .05 were considered to be statistically significant. Unadjusted relative risks (RRs) for the adverse outcomes in oral and parenteral cephalosporins were computed from 2 3 2 tables. Adjusted RRs were computed with Poisson regression models adjusting for age, sex, and cephalosporin generation. The primary outcome measures were new cephalosporin allergy reports within 30 days and incidence rates of serious cephalosporin-associated ADRs, including physician-documented anaphylaxis with the first-course exposure; physician-documented SCARs, hemolytic anemia, or nephropathy within 30 days; C difficile infection within 90 days; and all-cause death within 1 day.

RESULTS There were 3,999,290 total unique individuals with at least 1 health plan visit during the 3-year study period (from January 1, 2010, through December 31, 2012). There were 462,225 individuals (11.6% of total) with at least 1 hospitalization during the study interval. There were 2,848,001 individuals (71.2% of total) with only an outpatient visit during the study interval. There were 7767 individuals (0.19% of total) with only an e-visit during the study interval. The demographic characteristics of all health plan members, at their index visit during the study interval, who were hospitalized during the study interval were compared with those of health plan members who were not hospitalized (Table I). Female patients had almost twice as many baseline drug class allergies as did male patients. Hospitalized and nonhospitalized women were more likely than men to receive oral cephalosporins and had a higher baseline prevalence of penicillin or cephalosporin allergies. Penicillin allergy was noted on the day before the index visit in 6.3% of nonhospitalized individuals, 9.9% of hospitalized individuals, and 6.8% of all health plan members. A cephalosporin allergy was noted on the day before the index

MACY AND CONTRERAS 3

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE I. Demographic characteristics of hospitalized vs nonhospitalized health plan members

Characteristic

Numbers of individuals Age (y),* mean 6 SD Follow-up (mo),* mean 6 SD Maximum 5 36 mo Hospital days, mean 6 SD Drug class ‘‘allergies,’’* mean 6 SD Penicillin ‘‘allergy’’ history, n (%)* Cephalosporin ‘‘allergy’’ history, n (%)* Unique patients exposed to at least 1 oral cephalosporin course (% exposed) [total courses] Unique patients exposed to at least 1 parenteral cephalosporin course (% exposed) [total courses]

Nonhospitalized health plan members with at least 1 health care visit during the study interval (n 5 3,537,065)

Health plan members with at least 1 hospitalization during the study interval (n 5 462,225)

1,836,414 women 34.7 6 21.4 27.4 6 9.5

1,700,651 men 33.2 6 21.3 26.4 6 9.8

287,946 women 44.0 6 24.2 30.8 6 7.5

174,279 men 49.2 6 27.2 30.2 6 8.2

None 0.3 6 0.8 140,793 (7.7) 19,583 (1.1) 291,233 (15.9) [409,292]

None 0.2 6 0.5 83,499 (4.9) 8,667 (0.5) 189,958 (11.2) [243,773]

4.9 6 9.7 0.6 6 1.2 32,491 (11.3) 5,875 (2.0) 91,444 (31.8) [164,058]

6.2 6 11.9 0.4 6 0.8 13,372 (7.7) 1,780 (1.0) 49,821 (28.6) [84,785]

66,321 (3.6) [80,249]

57,054 (3.4) [68,732]

119,055 (41.3) [190,408]

84,437 (48.4) [148,241]

Data recorded at index visit. *At or from first health plan visit during the study interval (index visit).

TABLE II. Demographic characteristics of cohort members exposed to cephalosporins Characteristic

Men (n 5 330,269)

Women (n 5 489,855)

Age (y),* mean 6 SD Follow-up (mo),* mean 6 SD Maximum 5 36 mo Hospital days, mean 6 SD No. with at least 1 hospitalization

40.82 6 25.42 30.50 6 7.26 2.51 6 8.75 100,950

2.08 6 7.47 158,669

Outpatient, ED, and e-visits, mean 6 SD No. with at least 1 outpatient, ED, or e-visit Drug class ‘‘allergies,’’* mean 6 SD Maximum 5 24 drug class ‘‘allergies’’ Total cephalosporin exposures, mean 6 SD (quartiles) [range] Oral cephalosporin exposures, mean 6 SD (quartiles) [range] Parenteral cephalosporin exposures, mean 6 SD (quartiles) [range]

13.14 6 13.08 229,319 0.27 6 0.65

15.27 6 15.22 331,186 0.53 6 1.03

1.65 6 1.46 (1, 1, 2) [1-79] 0.99 6 1.05 (0, 1, 1) [0-77] 0.66 6 1.10 (0, 0, 1) [0-71]

40.81 6 23.64 31.15 6 6.81

1.72 6 1.55 (1, 1, 2) [1-63] 1.17 6 1.23 (1, 1, 1) [0-59] 0.55 6 0.99 (0, 0, 1) [0-37]

Data collected at index visit. ED, Emergency department. *At or from first health plan visit during the study interval.

visit for 0.8% of nonhospitalized individuals and 1.7% of hospitalized individuals. Parenteral cephalosporins were given to nonhospitalized individuals on only rare occasions and were more likely to be given to hospitalized men than women. There were 1,389,538 total courses of cephalosporins given: 901,908 oral (64.9%) and 487,630 parenteral (35.1%), given to 820,124 (20.5%) unique individuals during the study period. There were 622,456 (15.6%) individuals who received at least 1 oral cephalosporin, 326,867 (8.2%) who received at least 1 parenteral cephalosporin, and 129,199 (3.2%) who received at least 1 oral cephalosporin and at least 1 parenteral cephalosporin. The demographic characteristics of these patients are presented in Table II. Women had a higher rate of new cephalosporin allergy than did men. Cohort members who received a cephalosporin were older at their index visit than were health plan members who did not receive a cephalosporin (40.8 6 24.4 years vs 34.1 6 21.4 years), spent more days in the hospital (2.3 6 8.0 days vs 0.2 6 1.6 days), and had more baseline drug class allergies (0.46 6 0.95 allergies vs 0.25 6 0.67 allergies) (P < .0001 for all 3 comparisons). The route of administration of cephalosporins, the number of

individuals exposed, the total number of cephalosporin courses, new allergy reports, and serious ADRs are presented, by sex, in Table III. The specific cephalosporins used by each sex are listed in Tables E1 and E2 in this article’s Online Repository at www. jacionline.org. The courses of cephalosporins given to individuals with histories of drug allergies are presented, by sex, in Table IV along with reports of new cephalosporin allergies (per course) within 30 days, physician-documented anaphylaxis, nephropathy within 30 days, death within 1 day, and C difficile infection within 90 days. Individuals with any history of drug allergy received cephalosporin at an older mean age than did those without a history of allergy. This could be because older individuals were more likely to receive multiple cephalosporin courses and also more likely to have a drug allergy history. The specific route and generation of the cephalosporins given to these individuals are presented in Tables E3 and E4 in this article’s Online Repository at www.jacionline.org. A greater proportion of individuals with a history of penicillin allergy (1.13%; 95% CI, 1.07% to 1.19%) had a new report of cephalosporin allergy within 30 days of a cephalosporin course

4 MACY AND CONTRERAS

J ALLERGY CLIN IMMUNOL nnn 2014

TABLE III. Adverse reactions associated with cephalosporin use, by route Oral Adverse reaction

Men

Parenteral Women

Men

Women

Unique individuals 239,779 382,677 141,491 185,376 exposed Courses 328,558 573,350 216,973 270,657 New ‘‘allergy’’ reports/ 1605 (0.49) [0.46%-0.51%] 3351 (0.56) [0.54%-0.57%] 727 (0.34) [0.31%-0.36%] 1352 (0.50) [0.47%-0.53%] course within 30 d, n (%) [95% CI] Physician-documented 1 (0.0003) [0%-0.0009%] 4 (0.0007) [0.00001%-0.001%] 2 (0.0009) [0%-0.002%] 6 (0.002) [0.0004%-0.004%] anaphylaxis, n (%) [95% CI] All-cause death within 43 (0.01) [0.009%-0.02%] 32 (0.006) [0.004%-0.008%] 690 (0.32) [0.29%-0.34%] 682 (0.25) [0.23%-0.27%] 1 d, n (%) [95% CI] Nephropathy within 173 (0.05) [0.04%-0.06%] 136 (0.02) [0.02%-0.03%] 1,043 (0.48) [0.45%-0.51%] 796 (0.29) [0.27%-0.31%] 30 d, n (%) [95% CI] 961 (0.29) [0.27%-0.31%] 1771 (0.31) [0.29%-0.32%] 4402 (2.03) [1.97%-2.09%] 5502 (2.03) [1.98%-2.09%] C difficile infection within 90 d, n (%) [95% CI]

than did individuals with no reports of drug allergy (0.39%; 95% CI, 0.37% to 0.40%), individuals with cephalosporin allergy (0.70%; 95% CI, 0.50% to 0.91%), or individuals with allergy to another drug that was not a b-lactam (0.50%; 95% CI, 0.48% to 0.52%). There would be about 1 new cephalosporin allergy noted for every 135 individuals given cephalosporin who have penicillin allergy, compared with individuals with no drug allergies. Anaphylaxis was significantly overcoded, with only 13 of 21 events (61.9%) meeting the current working definition on the basis of the EHR review. The unconfirmed cases were mostly hives, with 1 case of erythema multiforme. Oral cephalosporins were less commonly associated with physician-documented anaphylaxis (5 of 901,908 courses; 95% CI, 1/1,428,571-1/ 96,154) than with parenteral cephalosporins (8 of 487,630 courses; 95% CI, 1/200,000-1/35,971). However, this difference did not quite reach statistical significance (P 5 .0761). Physiciandocumented anaphylaxis was 2.2-fold more common in women (10 of 844,007 cases; 95% CI, 1/222,222-1/52,110) than in men (3 of 545,531 courses; 95% CI, 0-1/85,324). However, this difference also did not reach statistical significance (P 5 .2709). There were 65,915 individuals with a history of a penicillin allergy who received a total of 127,125 courses of cephalosporins. There were 3 cases of cephalosporin-associated anaphylaxis in this group. Physician-documented cephalosporin-associated anaphylaxis was about 2.9-fold more common in individuals with histories of penicillin allergy than in individuals with no history of drug allergy (3 of 127,125 courses; 95% CI, 0-1/19,880 vs 7 of 845,923 courses; 95% CI, 1/467,290-1/69,396). This difference was not statistically significant (P 5 .1322). There were 3313 individuals with a history of a cephalosporin allergy; they received a total of 6404 courses of cephalosporin antibiotics, with no cases of cephalosporin-associated anaphylaxis. Individuals with allergies only to other drugs had a similar incidence of physician-documented cephalosporin-associated anaphylaxis (3 of 412,594 courses; 95% CI, 0-1/64,516) as did individuals with no drug allergies and individuals with penicillin allergy (P 5 1 and .1472, respectively). Female patients with oral cephalosporin-associated anaphylaxis were 12, 21, 26, and 72 years old, and the male patient was 49 years old. Female patients with parenteral

cephalosporin-associated anaphylaxis were 20, 33, 37, 41, 53, and 60 years old, and the male patients were 70 and 71 years old. There were only 3 total cases of a physician-diagnosed cephalosporin-associated SCAR (1 of 463,180 courses; 95% CI, 0-1/217,291); all these patients had SJS, identified in the study cohort. All cephalosporin-associated cases of SCAR were also associated with the use of another antibiotic at the time of the use of cephalosporin. The cases included a 53-year-old woman with ceftazidime- and vancomycin-associated SJS, an 18-year-old man with cephalexin- and cotrimoxazole-associated SJS, and a 20-year-old man with ceftriaxone- and azithromycin-associated SJS. No cases of cephalosporin-associated erythema multiforme major, toxic epidermal necrolysis, or drug reaction with eosinophilia and systemic symptoms were identified. There were 6601 cases (0.48%) of new-onset isolated eosinophilia of 10% or more, with absolute eosinophil counts of 500 or more/mL noted within 30 days of use of a cephalosporin. The top 3 cephalosporins associated with isolated eosinophilia were ceftriaxone (33.8% of the cases), cefazolin (23.2% of the cases), and cephalexin (18.8% of the cases). There were only 3 unique individuals with both a new-onset hemoglobin level of less than 8 g/dL and a lactate dehydrogenase level of more than 2000 U/L measured within 30 days of starting a cephalosporin course and without a pre-existing diagnosis of anemia, sepsis, or renal failure. The EHR review revealed 2 possible cases of cephalosporin-associated hemolytic anemia and 1 probable case. These cases are reported in detail in this article’s Online Repository at www.jacionline.org. There were 1658 individuals (0.2%) with new-onset serum creatinine levels of 3 or more mg/dL within 30 days of starting a cephalosporin course and no history of renal disease or increases in serum creatinine level. These individuals were exposed to a total of 2148 overlapping courses of cephalosporins. Among these 1658 individuals, none was exposed to an aminoglycoside within 1 week of the use of a cephalosporin. There were 877 patients (52.9%) who reduced their serum creatinine level to less than 1.3 within 6 months without any dialysis. There were 386 patients (23.3%) who died within 30 days of cephalosporin use. There were 278 patients (16.8%) who had dialysis within

MACY AND CONTRERAS 5

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE IV. Cephalosporins use, new ‘‘allergies,’’ and associated adverse reactions with history of drug ‘‘allergy’’ Characteristic

No history

Penicillin allergy

Cephalosporin allergy

Allergy to any other drug

Total male health plan members 1,614,620 98,195 10,702 154,293 Total female health plan members 1,616,868 177,088 26,308 313,734 Average age of male health plan members (y),* 32.8 6 21.7 42.5 6 23.5 39.6 6 25.1 51.0 6 21.9 mean 6 SD Average age of female health plan members (y),* 31.7 6 20.8 47.9 6 21.3 48.7 6 21.8 52.0 6 19.2 mean 6 SD Drug class ‘‘allergies’’ at index visit in men,* None 1.4 6 0.8 1.9 6 1.2 1.3 6 0.6 mean 6 SD Drug class ‘‘allergies’’ at index visit in women,* None 1.9 6 1.4 2.8 6 1.9 1.4 6 0.8 mean 6 SD Drug class ‘‘allergies’’ at end of study in men, 0.1 6 0.3 1.5 6 0.9 2.0 61.3 1.4 6 0.8 mean 6 SD Drug class ‘‘allergies’’ at end of study in women, 0.1 6 0.4 2.1 6 1.5 3.0 6 2.1 1.6 6 1.0 mean 6 SD Total no. of unique men exposed to 251,762 (15.6) [379,071] 21,957  (22.4) [40,483] 857  (8.0) [1,493] 62,559 (40.5) [124,930] cephalosporins, n (% exposed) [total courses] Total no. of unique women exposed to 311,643 (19.3) [466,852] 43,958à (24.8) [86,642] 2,456à (9.3) [4,911] 143,082 (45.6) [287,664] cephalosporins, n (% exposed) [total courses] Average age of men given cephalosporins (y),§ 41.0 6 26.1 42.1 6 30.9 53.9 6 28.2 59.4 6 21.5 mean 6 SD Average age of women given cephalosporins (y),§ 36.4 6 23.9 51.3 6 26.3 60.7 6 23.0 57.4 6 20.2 mean 6 SD Drug class allergies before use of cephalosporin None 1.7 6 1.1 2.8 6 1.8 1.5 6 0.8 in men,§ mean 6 SD Drug class allergies before use of cephalosporin None 2.6 6 1.9 4.5 6 2.7 1.8 6 1.2 in women,§ mean 6 SD New cephalosporin allergy within 30 d in men, 1343 (0.35) [0.34%-0.37%] 424 (1.05) [0.95%-1.15%] 3 (0.20) [0%-0.43%] 502 (0.40) [0.37%-0.44%] n (% per course) [95% CI] New cephalosporin allergy within 30 d in women, 1933 (0.41) [0.40%-0.43%] 1015 (1.17) [1.10%-1.24%] 42 (0.86) [0.60%-1.11%] 1559 (0.54) [0.52%-0.57%] n (% per course) [95% CI] Physician-documented anaphylaxis in men, n (%) 2 (0.0005) [0%-0.0013%] 1 (0.002) [0%-0.0073%] 0 (0) [0%-0%] 0 (0) [0%-0%] [95% CI] Physician-documented anaphylaxis in women, 5 (0.0001) [0.0001%-0.002%] 2 (0.0023) [0%-0.0055%] 0 (0) [0%-0%] 3 (0.001) [0%-0.002%] n (%) [95% CI] All-cause death within 1 d in men, n (%) [95% CI] 399 (0.11) [0.09%-0.12%] 76 (0.19) [0.15%-0.23%] 3 (0.20) [0%-0.43%] 256 (0.20) [0.18%-0.23%] All-cause death within 1 d in women, n (%) 250 (0.05) [0.05%-0.06%] 152 (0.18) [0.15%-0.20%] 12 (0.24) [0.11%-0.38%] 307 (0.11) [0.09%-0.12%] [95% CI] Nephropathy within 30 d in men, n (%) [95% CI] 759 (0.20) [0.19%-0.21%] 106 (0.26) [0.21%-0.31%] 6 (0.40) [0.08%-0.72%] 347 (0.28) [0.25%-0.31%] Nephropathy within 30 d in women, n (%) [95% CI] 411 (0.09) [0.08%-0.10%] 137 (0.16) [0.13%-0.19%] 6 (0.12) [0.02%-0.22%] 380 (0.13) [0.12%-0.15%] C difficile infection within 90 d in men, n (%) 2696 (0.71) [0.68%-0.74%] 560 (1.38) [1.27%-1.50%] 25 (1.67) [1.02%-2.33%] 2090 (1.67) [1.60%-1.74%] [95% CI] C difficile infection within 90 d in women, n (%) 2096 (0.45%) [0.43%-0.47%] 1243 (1.43%) [1.36%-1.51%] 157 (3.20%) [2.70%-3.69%] 3845 (1.34%) [1.29%-1.38%] [95% CI] *At or from first health plan visit during the study interval.  A total of 266 had both a penicillin and a cephalosporin allergy. àA total of 999 had both a penicillin and a cephalosporin allergy. §On the day before cephalosporin use.

6 months after a cephalosporin exposure and 2 patients (0.12%) who received kidney transplants before the end of the study period. Individuals with possible cephalosporin-associated nephropathy who died within 30 days tended to be older (66.9 6 16.4 years old vs 62.9 6 16.3 years old), men (59.6% vs 56.1% women), and hospitalized (88.1% vs 83.0% not hospitalized). The top 3 cephalosporins used by individuals with nephropathy-associated death within 30 days were ceftriaxone (by 50.8%), cefazolin (by 15.0%), and ceftazidime (by 15.0%). Among survivors, the top 3 cephalosporins used were ceftriaxone (by 48.6%), cefazolin (by 24.6%), and cephalexin (by 10.1%). Overall, parenteral use of cephalosporin was associated with rates of C difficile infection about 6-fold higher than in patients who received oral cephalosporins. Oral first-generation cephalosporins were associated with the lowest incidence of C difficile infection. Third- or higher generation parenteral

cephalosporins were associated with the highest incidence of C difficile infection—about 10-fold higher, or 3% per course, than were other cephalosporins (data not shown). There were 75 courses of oral cephalosporins given to patients who died from all causes within 1 day (1 of 12,025; 95% CI, 1/15,542-1/98). There were 1372 courses of parenteral cephalosporins given to patients who died from all causes within 1 day (1 of 355; 95% CI, 1/375-1/338) (P < .0001, compared with oral cephalosporins). The 1341 individuals who died from all causes within 1 day of starting a cephalosporin tended to be elderly (mean age, 73.0 6 17.0 years), men (51.2%), hospitalized (98.0%), and receiving a third- or higher generation parenteral cephalosporin (83.7%). None of these individuals had physician-diagnosed anaphylaxis. The immediate cause of death was not coded for 927 (69.1%). The 5 most common coded immediate causes of death within 1 day of starting a cephalosporin were pneumonia (81 cases, 6.0%), myocardial infarction

6 MACY AND CONTRERAS

(78 cases, 5.8%), lung cancer (57 cases, 4.3%), coronary artery disease (45 cases, 3.4%), and Alzheimer’s disease (39 cases, 2.9%). There were 166,848 (34.2%) courses of parenteral cephalosporins given on the same day as a surgery; these were presumed to be given for surgical prophylaxis. Parenteral cephalosporins, most commonly cefazolin (63.0%), ceftriaxone (29.4%), and ceftazidime (3.8%), were the most commonly given on the same day as a surgery. New cephalosporin allergies were reported to occur at 0.24%/course among individuals receiving a parenteral cephalosporin on the same day as surgery; this value is lower than the rate of 0.48%/course reported for individuals receiving a parenteral cephalosporin for presumed bacterial infections (P < .0001). The unadjusted RR for new report of cephalosporin-associated allergy within 30 days of administration was 1.29. This means that oral cephalosporins are 1.29-fold more likely than parental cephalosporins to result in a new report of allergy. However, parenteral cephalosporins were 2.9-fold more likely to result in anaphylaxis, 33.3-fold more likely to result in death within 1 day of administration, and 6.7-fold more likely to result in C difficile infection within 90 days than were oral cephalosporins. Nephropathy was equally likely to result from oral or parental cephalosporins. Age was significant only in the model used to compute the adjusted RR for C difficile infection. After adjusting for sex and cephalosporin generation, new allergy reports within 30 days (RR 5 1.92) were more likely with oral cephalosporins, death within 1 day (RR 5 0.08) was more likely with parenteral cephalosporins, and nephropathy (RR 5 1.0) was still equally likely with oral and parenteral cephalosporins. Anaphylaxis was 3.4-fold more likely to occur in patients receiving parenteral cephalosporins after adjusting for sex, and C difficile infection was 1.15-fold more likely in patients receiving parenteral cephalosporins, after adjusting for age.

DISCUSSION Oral and parenteral cephalosporins are widely and safely used in individuals with histories of allergies to penicillin and, surprisingly, other cephalosporins. During the study interval, about 24.4% of all health plan members with a history of penicillin allergy were treated with a cephalosporin. Unexpectedly, about 9.2% of all health plan members with a history of a cephalosporin allergy were treated with another cephalosporin and had a lower incidence of new cephalosporin allergy than did individuals with histories of penicillin allergy. Essentially no cephalosporin allergy tests or graded challenges were performed for health plan members during the study period although there is an active penicillin allergy testing program at the San Diego Medical Center.1 One strength of our study is that we collected data from all health plan members (about 1% of the US population) over a 3-year period. This makes our findings widely generalizable. A limitation of our study is that our results rely on the diagnoses made and the documentation provided by physicians, along with the drug allergy reports from all health care providers, including physicians, pharmacists, physician assistants, nurse practitioners, nurses, and medical assistants. We found that anaphylaxis was significantly overcoded. Anaphylaxis may also be undercoded, with true episodes of anaphylaxis assigned other codes such as those for wheezing,

J ALLERGY CLIN IMMUNOL nnn 2014

angioedema, hypotension, drug allergy, or adverse effects of medical or biological substances. There is no plausible explanation for why cephalosporin-associated anaphylaxis would be less likely to be correctly coded for individuals with parenteral exposures or those with histories of penicillin allergy. It was not possible to manually review all the cases that might have been incorrectly coded with our resources. In any case, if a cephalosporin-associated allergy or serious ADR occurs and is not recorded or is recorded incorrectly, the event will have no effect on health care utilization. Studies are underway to determine the exact causes of death within 1 day of cephalosporin use and to accurately identify incorrectly or uncoded cephalosporin-associated anaphylaxis. We found reports of new cephalosporin allergy to occur at a lower rate in men, and at a slightly lower rate in women, than in our previous study of a small group of outpatients; this study used a slightly different methodology.10 In that study, women were significantly more likely to have a new cephalosporin allergy entered into the EHR after each cephalosporin use. In our present study, we limited new cephalosporin allergy reports to those recorded within 30 days of the start of the cephalosporin course, to more accurately associate each entry with a specific course of cephalosporin. This difference in methodology resulted in lower rates than we previously reported.10 Unexpectedly, new cephalosporin allergies were significantly more common with oral than with parenteral exposure. IgE- or other immune-mediated mechanisms might mediate a small fraction of new reports of cephalosporin allergy we identified among individuals with histories of penicillin allergy. However, 95% to 98% of these individuals had negative results from penicillin skin tests or oral amoxicillin challenge,4 so anticephalosporin IgE cross-reaction or coproduction with antipenicillin IgE was unlikely to be responsible for most of the cases. We previously published data on the lack of higher rates of new reports of cephalosporin allergy among 118 penicillin skin-test positive individuals receiving 169 courses of cephalosporin, compared with 1566 penicillin skin-test negative individuals receiving 2485 cephalosporin courses, from January 1, 2000, through December 31, 2009.13 Physician-documented cephalosporin-associated anaphylaxis was about 3-fold more common per course of cephalosporin via parenteral, compared with oral, exposure. However, even in our large cohort, the P value (.0761) did not quite reach statistical significance. Our rates were lower than those reported in 2003 by Kaufman and Kelly14 for the International Collaborative Study of Severe Anaphylaxis, about 1 of 17,500 individuals given either oral or parenteral cephalosporins.14 The International Collaborative Study of Severe Anaphylaxis reviewed only the clinical outcomes of about 10% as many cephalosporin exposures and were unable to validate the cases by the EHR review as we did. In 2011, Tejedor Alonso et al15 reported anaphylaxis in about 1 of 26,000 parenteral exposures to cefazolin, cefotaxime, and ceftriaxone (4 cases in 104,350 people). In 2007, Mulla and Simon16 analyzed data on all cases of cephalosporin-associated anaphylaxis identified on the basis of coding in the state of Florida during 2001. They were able to identify only 17 cases from a population of just over 16,000,000 people. However, they were unable to perform the same in-depth analysis of individual medical records that we performed, or determine the total population exposure to cephalosporins.16 In 2007, Johannes et al17 studied

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

cases of anaphylaxis that resulted in emergency room visits or hospitalization and reported that anaphylaxis occurred for 1 of 49,000 cephalosporin exposures (6 of 295,548 exposures). However, they did not differentiate between oral and parenteral exposures. Even though physician-documented cephalosporinassociated anaphylaxis was more than twice as frequent in women than in men, this difference did not reach statistical significance, consistent with most other studies on anaphylaxis.18 Case reports of uniquely cephalosporin-associated SCARs are rare.19 Many of these patients were using other medications at the same time as cephalosporin. We were unable to identify any case of a cephalosporin-associated SCAR associated with exposure to only cephalosporin. Cephalosporins have been previously associated with nephrotoxicity, identified on the basis of new onset of high levels of serum creatinine, particularly in children.20,21 The combined use of cephalosporins and aminoglycosides has been proposed to increase the risk for nephrotoxicity, though we did not observe any such association.22 Cephalosporin-associated hemolytic anemia has been reported on rare occasions.23,24 We were able to identify only 1 probable case in almost 1.4 million cephalosporin exposures. Minimizing the use of higher generation, and particularly parenteral, cephalosporins, when possible, should help reduce the high rates of C difficile infection.25 Mitropoulos et al26 reviewed adverse events associated with the use of oral cephalosporins. They found that skin rashes and gastrointestinal problems were the most common ADRs for cephalosporin, many of which do not result in a new report of cephalosporin allergy. What is entered in the drug allergy field of a health record may have more to do with nonimmunologic factors such as a patient’s likelihood to report a reaction and the provider’s perception of the severity of the reaction, along with the perceived risk associated with re-exposure. Yoon et al27 recently reported that only 4 of 1421 individuals (0.28%) experienced acute itching or urticaria after receiving parenteral cephalosporins for surgical prophylaxis, similar to the proportion we observed (0.24%). Interestingly, the authors also performed cephalosporin skin tests on all participants and found that 74 of 1421 (5.2%) had a positive result (a 3 mm wheal). None of the subjects with positive results from the skin test reacted to subsequent parenteral exposure to cephalosporin. Furthermore, the 4 subjects who reacted to parenteral cephalosporins had negative results from the skin test. Cephalosporins are widely, safely, and appropriately used to treat individuals with a history of penicillin allergy. Our present findings confirm previous recommendations on the use of cephalosporins in individuals with a history of penicillin or other cephalosporin allergy.28 Physician-documented cephalosporinassociated anaphylaxis and uniquely cephalosporin-associated SCARs, though potentially life-threatening, are extremely rare. C difficile infection within 90 days of cephalosporin exposure is the most common serious ADR we identified, followed by nephropathy within 30 days and all-cause death within 1 day. More appropriate use of cephalosporins could help reduce serious cephalosporin-associated ADRs. Warnings against the administration of cephalosporins to patients with unconfirmed penicillin allergy should be removed from electronic medical record systems. The public would be better served by warnings that the use of cephalosporins, particularly third- or higher generation parenteral

MACY AND CONTRERAS 7

cephalosporins, is associated with an increased risk of C difficile infection within 90 days. Further research is required to determine the causes of death within 1 day of cephalosporin use and the mechanisms responsible for new-onset cephalosporinassociated nephropathy within 30 days. Clinical implications: New reports of cephalosporin ‘‘allergy’’ are more frequent for women than for men and are more common with oral than with parenteral exposure. Physiciandocumented cephalosporin-associated anaphylaxis is rare, even among individuals with a history of penicillin allergy. The most frequent serious cephalosporin-associated adverse drug reactions are C difficile infection within 90 days, nephropathy within 30 days, and all-cause death within 1 day, and all are unrelated to drug allergy history.

REFERENCES 1. Macy E, Contreras R. Healthcare utilization and serious infection prevalence associated with penicillin allergy in hospitalized patients: a cohort study. J Allergy Clin Immunol 2014;133:790-6. 2. Ren X, Liu D, Ding N, Huang K, Xiong Y, Du G, et al. Safety evaluation of cephalosporins based on utilization and adverse drug events: analysis of two databases in China. Expert Opin Drug Saf 2012;11:689-97. 3. Solensky R. Hypersensitivity reactions to beta-lactam antibiotics. Clin Rev Allergy Immunol 2003;24:201-19. 4. Macy E, Ngor E. Safely diagnosing clinically significant penicillin allergy using only penicilloyl-poly-lysine, penicillin, and oral amoxicillin. J Allergy Clin Immunol 2013;1:258-63. 5. Lagace-Wiens P, Rubinstein E. Adverse reactions to b-lactam antimicrobials. Expert Opin Drug Saf 2012;11:381-99. 6. Lee P, Shanson D. Results of a UK survey of fatal anaphylaxis after oral amoxicillin. J Antimicrob Chemother 2007;60:1172-3. 7. Van Der Klauw MM, Stricker BHCH, Herings RMC, Cost WS, Valkenburg HA, Wilson JHP. A population based case-cohort study of drug-induced anaphylaxis. Br J Clin Pharmacol 1993;35:400-8. 8. Hunziker TH, Kunzi UP, Braunschweig S, Zehnder D, Hoigne R. Comprehensive hospital drug monitoring (CHDM): adverse skin reactions, a 20-year survey. Allergy 1997;52:388-93. 9. Alanis A, Weinstein AJ. Adverse reactions associated with the use of oral penicillins and cephalosporins. Med Clin N Am 1983;67:113-29. 10. Macy E, Poon KYT. Self-reported antibiotic allergy incidence and prevalence: age and sex effects. Am J Med 2009;122:778, e1-7. 11. Macy E, Ho NJ. Multiple drug intolerance syndrome: prevalence, clinical characteristics, and management. Ann Allergy Asthma Immunol 2012;108:88-93. 12. Campbell RL, Hagan JB, Manivannan V, Decker WW, Kanthala AR, Bellolio MF, et al. Evaluation of National Institute of Allergy and Infectious Diseases/Food Allergy and Anaphylaxis Network criteria for the diagnosis of anaphylaxis in emergency department patients. J Allergy Clin Immunol 2012;129:748-52. 13. Macy E, Ho NJ. Adverse reactions associated with therapeutic antibiotic use after penicillin skin testing. Perm J 2011;15:31-7. 14. Kaufman DW, Kelly JP. for the International Collaborative Study of Severe Anaphylaxis. Risk of anaphylaxis in a hospital population in relation to the use of various drugs: an international study. Pharmacoepidemiol Drug Saf 2003;12: 195-202. 15. Tejedor Alonso MA, Moro Moro M, Esteban Hernandez J, Mugica Garcia MV, Vila Albelda C, Rosado Ingelmo A, et al. Incidence of anaphylaxis in hospitalized patients. Int Arch Allergy immunol 2011;156:212-20. 16. Mulla ZD, Simon MR. Hospitalizations for anaphylaxis in Florida: epidemiologic analysis of a population-based dataset. Int Arch Allergy immunol 2007;144: 128-36. 17. Johannes CB, Ziyadeh N, Seeger JD, Tucker E, Reiter C, Faich G. Incidence of allergic reactions associated with antibiotic use in a large, managed care organization. Drug Saf 2007;30:705-13. 18. Sheikh A, Alves B. Age, sex, geographical and socio-economic variations in admissions for anaphylaxis: analysis of four years of English hospital data. Clin Exp Allergy 2001;31:1571-6. 19. Grgurevic I, Pejsa V, Morovic-Vergles J, Dobric I, Gasparovic V, Tudoric N. Fatal toxic epidermal necrolysis and severe granulocytopenia following therapy with cefuroxime. Acta Dermatovenerol Croat 2008;16:133-7.

8 MACY AND CONTRERAS

20. Tong JE, Howell DN, Foreman JW. Drug-induced granulomatous interstitial nephritis in a pediatric patient. Pediatr Nephrol 2007;22:306-9. 21. Longstreth KL, Robbins SD, Smavatkul C, Doe NS. Cephalexin-induced acute tubular necrosis. Pharmacotherapy 2004;24:808-11. 22. Zhang Q, Matsumura Y, Teratani T, Yoshimoto S, Mineno T, Nakagawa K, et al. The application of an institutional data warehouse to the assessment of adverse drug reactions (ADRs): evaluation of aminoglycoside and cephalosporin associated nephrotoxicity. Methods Inf Med 2007;46:516-22. 23. Viraraghavan R, Chakravarty AG, Soreth J. Cefotetan-induced haemolytic anaemia, a review of 85 cases. Adv Drug React Toxicol Rev 2002; 21:101-7.

J ALLERGY CLIN IMMUNOL nnn 2014

24. Seltsam A, Salama A. Cetriaxone-induced immune haemolysis: two case reports and a concise review of the literature. Intensive Care Med 2000;26:1390-4. 25. Lode H. Safety and tolerability of commonly prescribed oral antibiotics for the treatment of respiratory tract infections. Am J Med 2010;123:s26-38. 26. Mitropoulos IF, Rotschafer JC, Rodvold KA. Adverse events associated with the use of oral cephalosporins/cephems. Diagn Microbiol Inf Dis 2007;57:67S-76. 27. Yoon SK, Park SY, Kim S, Lee T, Lee YS, Kwon HS, et al. Validation of the cephalosporin intradermal skin test for predicting immediate hypersensitivity: a prospective study with drug challenge. Allergy 2013;68:938-44. 28. Macy E, Ngor E. Recommendations for the management of beta-lactam intolerance. Clin Rev Allergy Immunol 2014;47:46-55.

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

METHODS A 4-year-old boy with newly diagnosed T-cell lymphoblastic lymphoma received ceftazidime the day after his diagnosis with lymphoma. His hemoglobin level was recorded at 7.5 g/dL 1 day later, and his lactate dehydrogenase level was recorded at 4774 U/L 3 days later. He started chemotherapy 4 days after his lymphoma diagnosis. Ceftazidime-associated hemolytic anemia was not suspected by his treating physicians. A 6-year-old girl was newly diagnosed with a B-cell acute lymphoblastic leukemia 1 day after receiving parenteral ceftazidime. Her hemoglobin level was recorded at 7.9 g/dL 6 days after she received the ceftazidime, and her lactate dehydrogenase level was recorded at 3752 U/L 1 day afterward. She

MACY AND CONTRERAS 8.e1

started chemotherapy 3 days after her leukemia diagnosis (4 days after receiving the ceftazidime). Ceftazidime-associated hemolytic anemia was not suspected by her treating physicians. A 37-year-old women underwent cardiac catheterization, receiving parenteral cefazolin for prophylaxis. She then developed what was reported to be ‘‘idiopathic hemolytic anemia’’ by her treating physicians. Her hemoglobin level was recorded a 7.6 g/dL 5 days later, and her lactate dehydrogenase level was recorded at 2447 U/L 3 days later. Cefazolinassociated hemolytic anemia was not suspected by her treating physicians. Her anemia resolved over the next month, and she had no further exposure to cefazolin.

8.e2 MACY AND CONTRERAS

J ALLERGY CLIN IMMUNOL nnn 2014

TABLE E1. Specific cephalosporins used and adverse reactions associated with cephalosporin in men Cephalosporin

Unique individuals Total New allergy reports Death Nephropathy C difficile infection exposed courses per course within 30 d Anaphylaxis within 1 d within 30 d within 90 d

First-generation oral

Cefadroxil Cephalexin

52 214,096

72 283,806

0 1187

0 1

0 32

0 159

0 812

2,030 7,353 4

2,468 9,131 5

6 37 0

0 0 0

0 11 0

2 11 0

3 104 0

22,231 992

31,947 1,129

374 1

0 0

0 0

0 1

32 10

53,862

63,915

149

0

70

242

954

684 388 28,876

712 824 35,278

3 0 114

0 0 0

2 0 19

6 4 43

20 8 230

1,679 2,755 8,561 76,685

2,008 3,149 11,328 99,759

24 16 63 358

0 0 1 1

29 21 210 339

19 23 89 617

175 65 902 2,048

Second-generation oral

Cefaclor Cefpodoxime Cefprozil Third- or higher generation oral

Cefdinir Cefixime First-generation parenteral

Cefazolin Second-generation parenteral

Cefotetan Cefoxitin Cefuroxime Third- or higher generation parenteral

Cefepime Cefotaxime Ceftazidime Ceftriaxone

MACY AND CONTRERAS 8.e3

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE E2. Cephalosporins used and adverse reactions associated with cephalosporin in women

Cephalosporin

First-generation oral Cefadroxil Cephalexin Second-generation oral Cefaclor Cefpodoxime Cefprozil Third- or higher generation oral Cefdinir Cefixime First-generation parenteral Cefazolin Second-generation parenteral Cefotetan Cefoxitin Cefuroxime Third- or higher generation parenteral Cefepime Cefotaxime Ceftazidime Ceftriaxone Ceftaroline

Unique individuals exposed

Total courses

New allergy reports per course within 30 d

Anaphylaxis

Death within 1 d

Nephropathy within 30 d

C difficile infection within 90 d

267 358,750

310 527,594

0 2,925

0 3

0 22

1 122

0 1,517

3,239 9,955 4

3,991 12,407 5

29 87 0

0 0 0

1 8 0

1 12 0

18 199 0

19,954 1,102

27,794 1,249

303 7

1 0

0 1

0 0

32 5

78,459

89,503

226

3

60

141

1,051

837 216 38,605

856 322 47,558

2 0 344

0 0 0

0 0 12

6 1 25

19 8 307

1,581 2,347 8,344 90,455 1

1,880 2,665 10,780 117,092 1

29 14 87 650 0

0 0 0 3 0

43 16 202 349 0

11 10 87 515 0

144 60 912 3,001 0

8.e4 MACY AND CONTRERAS

J ALLERGY CLIN IMMUNOL nnn 2014

TABLE E3. Oral and parenteral cephalosporins, by generation, given to men and histories of drug allergy Cephalosporin

Oral first generation, n (%) Oral second generation, n (%) _ third generation, n (%) Oral > Parenteral first generation, n (%) Parenteral second generation, n (%) _ third generation, n (%) Parenteral > Supplemental data to Table III.

No history of drug allergy (n 5 251,762)

165,169 5,446 17,317 37,209 18,645 58,562

(65.6) (2.2) (6.9) (14.8) (7.4) (23.3)

Penicillin allergy (n 5 21,957)

11,474 864 4,956 2,382 2,660 5,572

(52.3) (3.9) (22.6) (10.9) (12.1) (25.4)

Cephalosporin allergy (n 5 857)

304 42 66 143 82 380

(35.5) (4.9) (7.7) (16.7) (9.6) (44.3)

Any other drug allergy (n 5 62,559)

39,395 3,056 1,526 14,737 8,805 18,692

(63.0) (4.9) (2.4) (23.6) (14.1) (29.9)

MACY AND CONTRERAS 8.e5

J ALLERGY CLIN IMMUNOL VOLUME nnn, NUMBER nn

TABLE E4. Oral and parenteral cephalosporins, by generation, given to women and histories of drug allergy Cephalosporin

Oral first-generation, n (%) Oral second-generation, n (%) _ third-generation, n (%) Oral > Parenteral first generation, n (%) Parenteral second generation, n (%) _ third generation, n (%) Parenteral > Supplemental data to Table III.

No history of drug allergy (n 5 311,643)

229,250 5,206 15,640 45,182 17,104 53,940

(73.6) (1.7) (5.0) (14.5) (5.5) (17.3)

Penicillin allergy (n 5 43,958)

28,137 2,096 4,398 5,625 5,852 10,170

(64.0) (4.8) (10.0) (12.8) (13.3) (23.1)

Cephalosporin allergy (n 5 2,456)

1,179 198 76 417 274 930

(48.0) (8.1) (3.1) (17.0) (11.2) (37.9)

Any other drug allergy (n 5 143,082)

105,820 5,780 1,386 27,997 16,776 31,881

(74.0) (4.0) (1.0) (19.6) (11.7) (22.3)