Clinical efficacy of virtual reality for acute procedural pain management: A systematic review and meta-analysis

Background: Acutely painful procedures are commonplace. Current approaches to pain most often involve pharmacotherapy, however, there is interest in virtual reality (VR) as a non-pharmacological alternative. A methodologically rigorous systematic review and meta-analysis is lacking. Methods: Following PRISMA guidelines, we searched the Cochrane Library, Ovid MEDLINE, Embase, CINAHL, ERIC, NIHR Centre for Review and Dissemination, Proquest, the System for Information on Grey Literature in Europe and the WHO International Clinical Trials Registry Platform from inception to 5 November 2017. Included studies were randomised with an experimental trial design, included a non-VR control group and examined the efficacy of VR with regards to an acutely painful clinical intervention. Bias was assessed along Cochrane guidelines, with performance bias not assessed due to the non-blindable nature of VR. We extracted summary data for maximal pain score and used standard mean difference DerSimonian-Laird random-effects meta-analysis (RevMan 5.3). This review was prospectively registered (PROSPERO CRD42017058204). Findings: Of the 12,450 studies identified, 20 studies were eligible for the systematic review. No trials reported in sufficient detail to judge their risk of bias, and 10 studies were at high risk of bias in at least one domain. 16 studies (9 randomised controlled trials, 7 crossover studies) examining 656 individuals were included in quantitative synthesis. Pain scales were heterogenous, but mostly employed 100-point scales. Across all trials, meta-analysis was suggestive of a -0.49 (95%CI -0.83 to -0.41, p = 0.006) standardised mean difference reduction in pain score with VR. However there was a high degree of statistical heterogeneity (χ2 p