Nicotine replacement therapy is widely used in critically ill smokers and its effect on delirium, mortality and duration of intensive care unit (ICU) admission is unknown. The aims of this review were to determine whether the management of nicotine withdrawal with nicotine replacement therapy reduces delirium, mortality or length of stay in critically ill smokers in ICU. The primary outcome was incidence of author-defined ICU delirium. Secondary outcomes were ICU or hospital mortality, ICU-free days at day 28, and ICU or hospital length of stay. We conducted a systematic review and meta-analysis of the data sources MEDLINE, EMBASE, CINAHL, and the Cochrane Database of Systematic Reviews for randomised controlled trials and observational studies. Clinical trials, observational studies and systematic reviews comparing nicotine replacement therapy with placebo or no treatment were included. Case reports, case series, non-systematic reviews and studies that involved children were excluded. Eight studies were eligible (n=2,636) for inclusion in the data synthesis. In a meta-analysis of observational studies, nicotine replacement therapy was associated with increased delirium (three studies; n=908; I2=0%; finite element method: odds ratio 4.03 [95% confidence interval 2.64, 6.15]; P <0.001). There was no difference in ICU mortality (three studies; n=1,309; P=0.10, I2=44%; finite element method: odds ratio 0.58; 95% confidence intervals 0.31-1.10) and hospital mortality or 28-day ICU-free days. In the absence of high-quality data, nicotine replacement therapy cannot currently be recommended for routine use to prevent delirium or to reduce hospital or ICU mortality in critically ill smokers.