By Richard R. Watkins, MD, MS, FACP, FIDSA

Associate Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports that he has received research support from Allergan.

SYNOPSIS: A retrospective study found that among 1,488 hospitalized patients who received an antibiotic, 298 experienced at least one antibiotic-associated adverse drug event. Furthermore, 287 of the antibiotic regimens were not clinically indicated, and 56 of these were associated with an adverse drug event.

SOURCE: Tamma PD, Avdic E, Li DX, Cosgrove SE. Association of adverse events with antibiotic use in hospitalized patients. JAMA Intern Med 2017;177:1308-1315.

Antibiotics have saved countless lives and momentously improved human health since their use in clinical practice began more than 70 years ago. Yet, these miracle drugs have downsides, including adverse drug events (ADEs), organ toxicities, and promotion of antimicrobial-resistant pathogens. ADEs associated with antibiotics include allergic reactions, which can range from mild (e.g., rash or pruritus) to life-threatening (e.g., anaphylaxis), as well as the development of Clostridium difficile infection (CDI). Tamma et al sought to determine the incidence of antibiotic-associated ADEs in a cohort of adult inpatients at Johns Hopkins Hospital. The study was a retrospective analysis that included all patients ≥ 18 years of age admitted to the general medical service and who received antibiotics for at least 24 hours between September 2013 and June 2014. Patients were excluded if they received antibiotics for prophylaxis, inhaled or topical antibiotics, anti-tuberculosis antibiotics, and antibiotics for noninfectious indications. Both inpatient and outpatient medical records were reviewed to obtain follow-up data about ADEs. The investigators followed the patients’ clinical course for ADEs up to 90 days after the first day of antibiotic administration for the development of CDI and multidrug-resistant organism (MDRO) infections and up to 30 days for other ADEs (e.g., tendinitis, gastrointestinal, renal, hepatic, dermatologic, cardiac, neurologic, or hematologic toxicities). At least two infectious disease physicians or pharmacists decided on the association between the antibiotic received and the subsequent ADE.

Of 5,579 patients admitted to the medical service during the study interval, 1,488 received antibiotics for at least 24 hours and were included in the analysis. There were 324 unique ADEs, and 298 of the patients experienced at least one antibiotic-associated ADE. The most frequently prescribed antibiotics were third-generation cephalosporins (607), parenteral vancomycin (544), and cefepime (414) of regimens. Every additional 10 days of antibiotic therapy conferred a 3% increased risk of an ADE. There were 236 ADEs that occurred during the hospitalization, while 88 occurred after discharge, which included 11 cases of CDI and 44 MDRO infections.

The researchers determined that 287 of the prescribed antibiotic courses were not clinically indicated, such as treating asymptomatic bacteriuria or noninfectious respiratory conditions like exacerbations of congestive heart failure. Notably, of the nonindicated regimens, 20% were associated with an ADE. The most common ADEs were gastrointestinal, renal, and hematologic abnormalities. Aminoglycosides, trimethoprim-sulfamethoxazole, and parenteral vancomycin were the most frequent agents associated with renal injury. Two patients experienced QTc prolongation; one had received azithromycin and the other had received ciprofloxacin. Seven patients who received cefepime developed neurological side effects, including encephalopathy and seizures.

The rate of CDI was 3.9 (95% confidence interval [CI], 3.0-5.2) per 10,000 person-days for patients receiving antibiotics, which corresponded to 54 of the study patients. The antibiotics most commonly associated with CDI were third-generation cephalosporins (present in 52% of regimens preceding CDI), cefepime (present in 48% of regimens preceding CDI), and fluoroquinolones (present in 35% of regimens preceding CDI). Finally, 314 of the 324 ADEs were considered significant by the investigators, who defined this as resulting in a new hospitalization (n = 10), prolonged hospitalization (n = 77), additional office or ED visits (n = 29), and additional testing (n = 198). No deaths occurred as a result of an ADE.


The most important and alarming findings in the study by Tamma et al were that 27% of patients admitted to the medical service at Johns Hopkins Hospital received an antibiotic for at least 24 hours and 20% of them developed an antibiotic-associated ADE. While the specific ADEs were not novel and have been well described (e.g., nephrotoxicity from aminoglycosides and vancomycin), what is interesting is that the study provides quantitative data that can be used to estimate the risk of ADEs from antibiotics. That 19% of the prescribed antibiotic regimens were not necessary and were associated with a significant number of ADEs underscores the need for vigilant antibiotic stewardship. One of the key roles of antibiotic stewardship programs must be to educate all healthcare providers that antibiotics carry significant risk for ADEs. The same holds true when physicians discuss antibiotics with patients, especially when the decision has been made to stop or withhold antibiotics when they are not indicated.

There were a few limitations to the study. First, the patients had been referred to a large tertiary academic medical center and tended to be sicker and present with more underlying comorbidities than patients at other institutions. This likely made them more susceptible to antibiotic ADEs. Second, the hospital maintained an antibiotic stewardship program that was active during the study period, which probably caused a reduction in antibiotic-associated ADEs. Thus, the incidence of ADEs in institutions without an antibiotic stewardship program might be higher than the 20% rate reported by Tamma et al. Finally, accurate estimates of some infrequently prescribed antibiotics (e.g., tigecycline, ceftaroline, penicillin) could not be calculated.

Antibiotic-associated ADEs are common and produce significant consequences, including higher healthcare costs, longer length of stay, and harmful toxicities. Therefore, using antibiotics judiciously must be an essential goal for all conscientious physicians. The prophetic advice of Hippocrates to “do no harm” remains as relevant today as it was in 400 B.C.