Objective: A systematic review and meta-analysis to assess the efficacy of earplugs as an ICU strategy for reducing delirium.
Data sources: MEDLINE, EMBASE, and the Cochrane Central Register of controlled trials were searched using the terms "intensive care," "critical care," "earplugs," "sleep," "sleep disorders," and "delirium."
Study selection: Intervention studies (randomized or nonrandomized) assessing the efficacy of earplugs as a sleep hygiene strategy in patients admitted to a critical care environment were included. Studies were excluded if they included only healthy volunteers, did not report any outcomes of interest, did not contain an intervention group of interest, were crossover studies, or were only published in abstract form.
Data extraction: Nine studies published between 2009 and 2015, including 1,455 participants, fulfilled the eligibility criteria and were included in the systematic review. Studies included earplugs as an isolated intervention (n = 3), or as part of a bundle with eye shades (n = 2), or earplugs, eye shades, and additional sleep noise abatement strategies (n = 4). The risk of bias was high for all studies.
Data synthesis: Five studies comprising 832 participants reported incident delirium. Earplug placement was associated with a relative risk of delirium of 0.59 (95% CI, 0.44-0.78) and no significant heterogeneity between the studies (I, 39%; p = 0.16). Hospital mortality was reported in four studies (n = 481) and was associated with a relative risk of 0.77 (95% CI, 0.54-1.11; I, 0%; p < 0.001). Compliance with the placement of earplugs was reported in six studies (n = 681). The mean per-patient noncompliance was 13.1% (95% CI, 7.8-25.4) of those assigned to receive earplugs.
Conclusions: Placement of earplugs in patients admitted to the ICU, either in isolation or as part of a bundle of sleep hygiene improvement, is associated with a significant reduction in risk of delirium. The potential effect of cointerventions and the optimal strategy for improving sleep hygiene and associated effect on patient-centered outcomes remains uncertain.