Using systematic reviews to study adverse drug reactions: The example of heparin-induced thrombocytopenia

Article type
Authors
Junqueira D1, Carvalho M2, Perini E1
1Centre of Drug Studies, School of Pharmacy, Federal University of Minas Gerais, Brazil
2Department of Clinical and Toxicological Analysis, School of Pharmacy, Federal University of Minas Gerais, Brazil
Abstract
Background: Methodology to systematically review drug efficacy based on randomized controlled trials (RCT) is well-established. Nevertheless, evidence-based healthcare may be supported by balanced evidence on benefits and harms of healthcare interventions. Development of standard recommendations to systematically review harm effect with the same emphasis on treatment benefit is of remarkable importance.

Objectives: To describe the experience of conducting a systematic review for an adverse drug reaction (ADR).

Methods: We conducted a systematic review based on standard methodology drawn on the Cochrane Handbook for Systematic Reviews of Interventions. The review was adapted to evaluate evidence about the risk of developing an ADR, heparin-induced thrombocytopenia (HIT). HIT is a paradoxical and unpredictable reaction diagnosed by clinical and laboratory assessment.

Results: The review question focussed on a specific ADR in a selected, susceptible population. This was a pragmatic decision mostly concernedwith the quality evaluation of the ADR and impact of the clinical decision. We considered both blinded and unblinded RCTs for inclusion, and we did not limit our search strategy to 'adverse effects'.. Search strategies considered the intervention (anticoagulant) and the outcome (HIT). The quality assessment applied the Handbook directions, but special attention was paid to the appropriateness of HIT diagnosis according to its clinical and serological profile. We included assessment of adequacy of ADR monitoring as an additional bias criterion. Only one study was included. Most studies were excluded because the definition of HIT or diagnosis process was inappropriate.

Conclusions: The few studies included in the review illustratethe limitations of RCT to evaluate unpredictable adverse events induced by drug use, such as HIT. Characteristics of some ADRs, together with poor quality reporting, analysis and discussion, may justify a systematic review focused on specific adverse outcomes and inclusion of observational studies.