Summary
Clinical question
Is monitoring of liver function, lipids and full blood count necessary in healthy people taking isotretinoin?
Background
Routine blood testing was recommended in the original licence for Roaccutane™ (isotretinoin) in 1983. In recent years, less frequent monitoring has been suggested by various authors.
Data sources
We performed four individual systematic searches of the MEDLINE database, via PubMed, from origin to 2 May 2021, supplemented by a hand search of all references in the identified papers.
Study selection
Inclusion criteria were any description of clinical symptoms, laboratory abnormalities and/or physical findings, and any paper that explicitly described the patients as asymptomatic, during treatment with oral isotretinoin.
Data extraction
Two independent reviewers (J.A. and D.J.) assessed articles for eligibility of inclusion. Evaluation of the data was done also by two of the authors (A.A., D.J. and J.A.) for each section, with the aim to use the presented evidence including guidelines, databases, case series, case reports, cohort studies and randomized clinical trials to delineate the clinical presentation and frequency of adverse events that might be amenable to laboratory monitoring.
Results
We identified 407 papers in our searches and reviewed 125 papers in four sections. Overall, reported adverse events were very rare (< 1 in 10 000) and were either idiosyncratic or not preventable by monitoring, accompanied by symptoms, or seen in identifiable predisposed individuals who might benefit from monitoring because of pre‐existing conditions.
Recommendation for clinical care
We could not find evidence to support the benefit of monitoring to detect adverse events. We suggest that in healthy young people laboratory monitoring for oral isotretinoin is unnecessary and risks detecting nonserious biochemical abnormalities. However, we recognize that new information about adverse events may change that recommendation.