Lara C. Pullen, PhD
November 19, 2014
The adverse effects of amoxicillin appear to be underreported in clinical trials. In particular, treatment with amoxicillin-clavulanic acid commonly results in diarrhea, and treatment with amoxicillin-clavulanic acid or amoxicillin commonly results in candidiasis.
Malcom Gillies, PhD, from NPS Medicine Wise, Sydney, New South Wales, Australia, and colleagues published their systemic review and meta-analysis of randomized trials onlineNovember 17 in the Canadian Medical Association Journal. The review was intended to evaluate clinical studies and inform clinicians about harms from amoxicillin, thereby enabling physicians to make more informed decisions when prescribing this common antibiotic.
The team focused on 45 clinical trials (27 studying amoxicillin, 17 studying amoxicillin-clavulanic acid, and 1 studying both). Only 25 trials provided data that allowed the authors to evaluate harms. The authors conclude from this that adverse events were underreported in the medical publications.
The authors found that 10% of courses of treatment with amoxicillin were associated with diarrhea. They calculated an odds ratio for diarrhea of 3.30 (95% confidence interval, 2.23 – 4.87) for patients receiving amoxicillin-clavulanic acid compared with control patients. The odds ratio for candidiasis was significantly higher, at 7.77 (95% confidence interval, 2.23 – 27.11).
“Reported harms were fewer than we expected from clinical anecdotal experience and observationally derived data, which have primarily reported common harms as rashes (at rates of 5%–8% of those treated and even higher, up to 20%, among those with mononucleosis treated with amoxicillin) and gastrointestinal disturbance. Some standard textbooks do not report candidiasis. At least 1 case–control study found a relative risk of 7 for thrush after therapy with amoxicillin or amoxicillin–clavulanic acid,” they write.
The authors emphasize that their finding is likely not unique to amoxicillin. Genuine adverse events are often underreported in clinical trials.
The underreporting may be a result of poor monitoring, missing data, or lack of clear case definitions. Whatever the cause, the underreporting flows into systematic reviews and treatment guidelines.
In an accompanying editorial, Yoon Kong Loke, MBBS, MD, and Katharina Mattishent, MBBS, from the University of East Anglia in the United Kingdom, explain that the risk for bias toward the null should be explicitly assessed in systematic reviews. Such an approach will discourage the false sense of security that occurs when a drug is incorrectly declared safe or not significantly different from a comparator.
Dr Loke and Dr Mattishent also point out that absence of evidence of harm is not the same as evidence of absence of harm.
They suggest that quality assessment tools for adverse advents be incorporated into clinical trials to submit the claim of “no significant harm” to the same scrutiny that efficacy claims receive. Patients and physicians would also benefit from unrestricted access to complete trial data that include information on efficacy as well as adverse events.
This study was conducted as a project of the Centre for Research Excellence in Minimising Antibiotic Resistance in Acute Respiratory Infections, with funding for the centre provided by the National Health and Medical Research Council of Australia. NPS MedicineWise is an independent nonprofit organization funded by the Australian Government’s Department of Health. One coauthor reports receiving personal fees from BMJ Books and Elsevier. The other authors and Dr Loke and Dr Katharina have disclosed no relevant financial relationships.
CMAJ. Published online November 17, 2014. Article full text