Aim To evaluate effectiveness and safety of midazolam as a sedative agent in children undergoing imaging procedures.

Methods A systematic literature review on MEDLINE (1948–September 2014), EMBASE (1980–September 2014), International Pharmaceutical Abstracts (1970 to September 2014) and PubMed database (until September 2014) in all languages. Studies evaluating midazolam use in children undergoing imaging procedural sedation were included if they reported efficacy and/or safety outcomes.

Results 25 studies, in 6145 patients, evaluated adverse events (AEs). Dosage and route ranged from 0.1 to 0.6 mg/kg for intravenous route, 0.15 to 0.45 mg/kg for intranasal, 0.5 to 0.6 mg/kg for oral and 0.3 to 1 mg/kg for rectal administration. The most common AE was hypoxia (74), in 3.6 per 100 patients. Cases were mild (SpO2 90–95%) in 42 (2.1% incidence) and moderate (SpO2<90%) in 32 cases (1.5%), with no severe cases. All were completely reversible after simple airway manoeuvres. Vomiting was the second most frequently, 1 per 100 patients. Two serious AEs occurred, both myoclonic-like movements of upper and lower extremities requiring medical interventions and hospitalisation. 24 studies, in 3717 patients, evaluated efficacy. Procedural success rates ranged from 0% to 100%, median 86%. Success was higher for shorter procedures; CT scan (median 86.7%) versus for MRI (median 59%).

Conclusions Midazolam for imaging procedural sedation has a low incidence of AEs, the occurrence of hypoxia emphasises the importance of monitoring during sedation. The success rate is variable, with shorter procedures seeing better results.