Common Antibiotics Linked With Serious Skin Reactions in Older Adults

Commonly prescribed oral antibiotics were linked to increased odds of serious drug-related skin rashes that resulted in emergency department (ED) visits or hospitalizations in adults over 65 years of age, according to a Canadian case-control study.

After multivariable adjustment, the odds of developing serious cutaneous adverse drug reactions with commonly used antibiotics were two to nearly three times greater, when compared with reference macrolides:

  • Sulfonamides: aOR 2.9 (95% CI 2.7-3.1)
  • Cephalosporins: aOR 2.6 (95% CI 2.5-2.8)
  • Nitrofurantoin: aOR 2.2 (95% CI 2.1-2.4)

Penicillins (aOR 1.4, 95% CI 1.3-1.5) and fluoroquinolones (aOR 1.3, 95% CI 1.2-1.4) were also linked to increased odds of serious cutaneous drug reactions when compared with macrolides, David N. Juurlink, MD, PhD, of the University of Toronto, and colleagues reported in JAMA.

“While we have suspected not all antibiotics pose the same risk of severe drug rash, our study confirmed this speculation,” co-author Erika Lee, MD, MSc, also of the University of Toronto, told MedPage Today. “Sulfonamide antibiotics and cephalosporins pose the highest risk. Our study adds another reason for judicious antibiotic prescription in the community.”

The association between serious drug rashes and nitrofurantoin was unexpected, Lee and co-authors pointed out, noting that nitrofurantoin is a common first-line treatment for uncomplicated urinary tract infections in older adults.

“The take home message [for physicians] remains using antibiotics judiciously, using the narrowest spectrum antibiotic recommended for the bacterial infection present, using the lowest therapeutic dose needed, and continuing therapy for the shortest length of time needed,” Eric Macy, MD, MS, an allergy and immunology specialist at Kaiser Permanente in San Diego, told MedPage Today. Macy was not associated with the study.

“Avoiding sulfonamides and cephalosporins — particularly third generation cephalosporins — when they are not the antibiotic of choice is prudent, based on the result of this paper and others,” he added.

Serious cutaneous adverse drug reactions ranged from drug rashes to Stevens-Johnson syndrome (SJS)/ toxic epidermal necrolysis (TEN), the latter of which has a fatality rate of 20% to 40%.

Roughly two hospital visits for cutaneous adverse drug reactions occurred for every 1,000 antibiotic prescriptions, but most drug reactions did not lead to hospitalization, the authors pointed out. ED visits or hospitalization rates were highest for cephalosporins at 4.92 per 1,000 prescriptions (95% CI 4.86-4.99) and sulfonamides at 3.22 per 1,000 prescriptions (95% CI 3.15-3.28). Median length of hospital stay was 6 days (IQR 3-13 days).

About one in eight case patients presenting at the ED were hospitalized for cutaneous adverse drug reactions. Of these, 9.6% required transfer to critical care, and 5.3% died while hospitalized.

“Antibiotics are really important when they’re needed, but they’re not just benign entities that we just apply liberally,” Elizabeth Phillips, MD, of Vanderbilt University Medical Center in Nashville, Tennessee, who was not associated with the study, told MedPage Today. “We have to intentionally think about what we’re doing. This has been a long-standing problem in the elderly.”

Macrolides, which were used as the reference drug in the study, are known to have a very low risk of cutaneous toxicity, Phillips said. “However, they have their own baggage,” pointing to azithromycin’s association with a higher risk of cardiovascular death.

“In addition, most serious cutaneous adverse reactions are delayed reactions where the patient has to be on the antibiotic for several days to weeks before a reaction occurs,” she pointed out. “Many of these reactions may be avoided by shorter antibiotic courses,” she said, adding that recommended duration of antibiotic treatment for most infections is not evidence-based.

Researchers identified over 3.2 million adults ages 66 years or older in Ontario, Canada with an outpatient antibiotic prescription from 2002 to 2022. The study used ICD-10 codes to identify serious cutaneous drug reactions, which included SJS/TEN, allergic purpura, generalized skin eruptions, erythematous conditions, and eosinophilia, among other drug-related exanthems.

Among the older adults with antibiotic prescriptions, researchers found 21,758 ED visits or hospital admissions for cutaneous adverse drug reactions within 60 days of receiving outpatient antibiotics. These case patients were matched with up to four control patients (n=87,025) who had received an antibiotic prescription in the previous 60 days but had no hospital or ED encounter related to serious cutaneous drug reactions.

The most commonly prescribed antibiotics were penicillins accounting for approximately 29% of prescriptions, followed by cephalosporins at about 18%. Fluoroquinolones accounted for 16.5% of prescriptions and macrolides for 14.8%. Nitrofurantoin, sulfonamides, and less commonly prescribed antibiotics (grouped as “other”) each accounted for less than 10%.

The authors noted that they did not have access to detailed hospital records and relied on ICD-10 codes to identify serious antibiotic-related skin reactions. The study did not consider individual genetic susceptibility which may be a risk factor for drug reactions, nor was it able to identify use of nonprescription medications that may have contributed to skin reactions.

  • author['full_name']

    Katherine Kahn is a staff writer at MedPage Today, covering the infectious diseases beat. She has been a medical writer for over 15 years.

Disclosures

The study was funded by the Canadian Institute for Health Research.

Lee and Juurlink reported no conflicts of interest. One study author reported multiple ties to industry.

Macy and Phillips reported no conflicts of interest.

Primary Source

JAMA

Source Reference: Lee EY, et al “Oral antibiotics and risk of serious cutaneous adverse drug reactions” JAMA 2024; DOI: 10.1001/jama.2024.11437.

Please enable JavaScript to view the

comments powered by Disqus.