A systematic review of interventions to facilitate extubation in patients difficult-to-wean due to delirium, agitation, or anxiety and a meta-analysis of the effect of dexmedetomidine

Abstract

Background: Delirium, agitation, and anxiety may hinder weaning from mechanical ventilation and lead to increased morbidity and healthcare costs. The most appropriate clinical approach to weaning in these contexts remains unclear and challenging to clinicians. The objective of this systematic review was to identify effective and safe interventions to wean patients that are difficult-to-wean from mechanical ventilation due to delirium, agitation, or anxiety. Methods: A systematic review was performed using MEDLINE, EMBASE, and PubMed. Studies evaluating mechanically ventilated patients deemed difficult-to-wean due to delirium, agitation, or anxiety, and comparing the effects of an intervention with a comparator arm were sought. Time-to-extubation was the primary outcome while the secondary outcome was intensive care unit (ICU) length of stay. Results: From 10,860 studies identified, eight met the inclusion criteria: six studies assessed dexmedetomidine while the remaining two assessed loxapine and biofeedback. Pooled analysis of studies assessing dexmedetomidine showed reduced time-to-extubation (six studies, n = 303) by 10.9 hr compared with controls (95% confidence interval [CI], -15.7 to -6.1; I2 = 68%) and ICU length of stay (four studies, n = 191) by 2.6 days (95% CI, 1.9 to 3.3; I2 = 0%). Nevertheless, the evidence was deemed to be of low quality given the small sample sizes and high heterogeneity. Studies assessing other interventions did not identify improvements compared with controls. Safety assessment was globally poorly reported. Conclusions: This systematic review and meta-analysis provides low quality evidence to suggest the use of dexmedetomidine in patients deemed difficult-to-wean due to agitation, delirium, or anxiety. Insufficient evidence was found regarding other interventions to provide any recommendation. Trial registration: PROSPERO (CRD42016042528); registered 15 July, 2016.