Intravenous magnesium administration errors, attributing factors and associated respiratory or cardiopulmonary arrest in obstetric and non-obstetric patients - A systematic review.

Background: The primary objective of this systematic review was to investigate intravenous (IV) magnesium administration errors and associated adverse outcomes in hospital settings. The secondary objective was to identify contributory factors using the human factors and analysis classification system (HFACS) framework.

Methods: PubMed, Scopus and Google Scholar were searched using the systematic search protocol for the past five decades. Magnesium administration errors were included, provided the route of administration was IV, and the clinical outcomes were described. Reports of errors via other routes were excluded.

Results: The search identified 32 reports (31 obstetric and 15 non-obstetric patients). Errors occurred in wide-ranging locations. Most errors (36 of 46) occurred during maintenance dose infusion. For obstetric patients, errors occurred in patients with preeclampsia (n = 24) or preterm labour (n = 7) during varying periods of pregnancy. In four women, perimortem (2 patients) or within a few hours of cardiopulmonary resuscitation, lower segment caesarean sections were performed. Nearly two-thirds (63%) of patients developed either respiratory arrest only (9 obstetrics and four non-obstetric) or cardiopulmonary arrest (10 obstetrics and six non-obstetric). The permanent harm occurred in six patients (death -3 and vegetative state -3). Most contributory factors were skill-based errors, adverse mental state, communication and coordination of care, and deficiencies in magnesium-related processes.

Conclusions: The findings highlight the need for improved IV magnesium administration practices across wards and acute care settings. Standardisation of magnesium products, their labelling, preparation, as well as reliable monitoring processes during the therapeutic use of magnesium, are essential to prevent magnesium infusion errors.