Effectiveness of interventions to prevent pressure injury in adults admitted to acute hospital settings : A systematic review and meta-analysis of randomised controlled trials

Background
Hospital-acquired pressure injuries cause significant harm to afflicted individuals, and financially burden hospitals. Most pressure injuries are avoidable with the use of preventative interventions. However, within acute hospital settings the effectiveness of pressure injury preventative interventions, as demonstrated by high-level evidence, requires examination.

Objective
Analyse the effectiveness of interventions to prevent pressure injury in adults admitted to acute hospital settings.

Design
Systematic review and meta-analysis of randomised controlled trials.

Data sources
CINAHL, MEDLINE, Scopus, Web of Science and Embase were searched in May/June 2019. In April 2020, searches were updated to the end of 2019.

Methods
Randomised controlled trials which investigated the effectiveness of pressure injury preventative interventions on pressure injury incidence, within adults admitted to acute hospital settings, were included. Trials limited to pressure injury treatment or specialty areas, and non-English reports, were excluded. Screening, extraction and risk-of-bias assessment were undertaken independently by two reviewers, with a third as arbitrator. Included studies were grouped by intervention type. Studies were synthesised narratively, and meta-analysis was undertaken where study interventions were similar. Using a random-effects model, primary meta-analyses were undertaken using intention-to-treat data.

Results
Of 2000 records, 45 studies were included in the systematic review which investigated nine different intervention types: continence management, heel protection devices, medication, nutrition, positioning, prophylactic dressings, support surfaces, topical preparations and bundled interventions. All studies were judged to be at unclear or high risk-of-bias. Several meta-analyses were undertaken, pooled by intervention type. Most pooled samples were heterogeneous. Based on intention-to-treat data, only one intervention demonstrated a statistically significant effect: Australian medical sheepskin surfaces compared to other standard care surfaces (risk ratio 0.42, p = 0.006, I2 = 36%), but included studies were limited by bias and age. Following per protocol meta-analyses, only two intervention types demonstrated a significant effect: support surfaces (active versus other comparison [risk ratio = 0.54, p = 0.005, I2 = 43%] and standard surfaces [risk ratio = 0.31, p < 0.001, I2 = 0%]; and reactive versus other comparison surfaces [risk ratio = 0.53, p = 0.03, I2 = 64%]) and heel protection devices versus standard care (risk ratio = 0.38, p < 0.001, I2 = 36%).

Conclusions
Only one intervention was supported by intention-to-treat meta-analysis. Significantly, all trials were at unclear or high risk-of-bias; and there were several limitations regarding heterogeneity across trials and trial outcomes. Further large-scale, high-quality trials testing pressure injury preventative interventions are required to establish effectiveness within acute hospital settings. Attention should be paid to true intention-to-treat analysis, and acute and intensive care settings should be reported separately. PROSPERO registration number: CRD42019129556.