Abstract Rationale Anxiety is prevalent among critically ill patients and can complicate recovery and affect patient safety. Conventional pharmacologic therapies for anxiety, such as benzodiazepines, can increase the risk of delirium and respiratory depression. Inhalation of essential oils have emerged as potential therapies for anxiety with minimal side effects. While individual randomized controlled trials (RCTs) have evaluated specific oils, direct comparisons are limited, necessitating a network meta-analysis (NMA) to synthesize evidence and rank interventions for anxiety reduction. Methods The analysis followed PRISMA guidelines for systematic reviews and NMAs. We searched databases including PubMed, Embase, and Cochrane Library up to August 2025 for RCTs assessing essential oils or related interventions (e.g., Benson relaxation, reflexology) in critically ill adults with anxiety. Data on mean differences (MD) in anxiety scores, such as State-Trait Anxiety Inventory (STAI), were extracted. A random-effects model was applied. Risk of bias was assessed using the Cochrane Risk-of-Bias tool (RoB 2). MetaInsight v6.4.0 was used for data synthesis and the Confidence in Network Meta-Analysis (CINeMA) was used to evaluate the level of confidence. Results Seventeen RCTs involving 1,587 patients were identified, forming a connected network with 66 possible pairwise comparisons (17 with direct evidence). After excluding two RCTs with high risk of bias, 15 RCTs involving 1,395 patients were included for analysis. The average baseline state STAI score was 43.14. Despite the limited number of studies, geranium demonstrated the greatest reduction in anxiety (MD -32.9, 95% CrI -53.0 to -12.5), followed by citrus aurantium (MD -14.6, -25.6 to -3.35). Although lavender is one of the most widely used essential oils in clinical settings, it showed a modest and non-significant effect (MD -3.76, -11.3 to 3.76). SUCRA rankings identified geranium as most effective (100%), followed by citrus aurantium (90%), lavender-matricaria-citrus combination (75%), melissa (65%), lemon (55%), rose (45%), lavender (40%), bergamot (30%), reflexology (20%), Benson (10%), and control (0%). Node-splitting revealed no significant inconsistency (e.g., rose vs. lavender p = 0.736; citrus aurantium vs. lavender p = 0.140). Conclusion This analysis suggests that geranium and citrus aurantium outperform other essential oils in reducing anxiety among critically ill patients. Larger, high-quality RCTs are needed to establish the role of these two oils in anxiety reduction across different ICU settings. The evidence for lavender remains inconclusive, warranting further research. Given their potential benefits and minimal side effects, essential oils may serve as valuable therapies for anxiety management in critically ill patients. This abstract is funded by: None
Mehta et al. (Fri,) studied this question.