Annabel Levido

The University of Queensland's institutional repository, UQ eSpace, aims to create global visibility and accessibility of UQ’s scholarly research.

Summary: Introduction: Although there is an extensive body of literature on complications associated with central venous catheters, less is known about arterial catheter‐related complications. This systematic review and meta‐analysis aimed to quantify the incidence of arterial catheter‐related complications and failure in intensive care. Methods: Databases were searched for relevant randomised controlled trials and cohort studies that reported on arterial catheter‐related complications and failure within the adult critical care setting. Two authors independently identified studies for full review, extracted data and completed quality assessments. Pooled estimates were calculated using random‐effects models with the Freeman‐Tukey double arcsine transformation. Certainty of evidence for each outcome was assessed using the GRADE framework. Results: Thirty‐nine studies (22 observational studies and 17 randomised controlled trials) comprising 19,018 arterial catheters, were included. The pooled proportion of all‐cause arterial catheter failure was 13.0% (95%CI 7.6–19.5%, moderate certainty). Non‐infectious arterial catheter failure occurred in 19.8% (95%CI 12.2–28.7%, moderate certainty) of arterial catheters. The pooled proportion of all‐cause catheter‐associated or related bloodstream infections was 1.3% (95%CI 0.7–2.1%, moderate certainty). Local infection occurred in 1.2% (95%CI 0.4–2.4%, high certainty). Reporting of definitions of bloodstream and local infections was limited. Incidence density could not be calculated for any outcomes due to limited reporting of total catheter days. Discussion: Arterial catheter failure occurs in up to one in five catheters in intensive care. Reducing failure rates and associated complications requires system‐wide interventions, clearer definitions of infection and standardised reporting to improve data synthesis and guide evidence‐based improvement.