The association of nicotine replacement therapy with mortality in a medical intensive care unit

Objective: Smokers admitted to the intensive care unit may receive nicotine replacement therapy to prevent withdrawal. However, the safety of nicotine replacement in the critically ill has not been studied. The objective of this study was to determine the impact of nicotine replacement on the outcome of critically ill patients. Design. Retrospective, case-control. Setting. The medical intensive care unit of a tertiary academic hospital. Patients: Patients who were active smokers at admission to the intensive care unit were included in the study. Those who received nicotine replacement therapy were considered as cases, and those who did not receive nicotine replacement were considered as controls. Interventions: None. Measurements and Main Results: For each of the 90 cases, one control smoker who did not receive nicotine replacement therapy was selected based on the severity of illness and then age. Outcome was measured by hospital mortality and 28-day intensive care unit-free days, defined as the number of days spent outside of intensive care or without mechanical ventilation by a living patient following admission to intensive care. The mean mortality rate predicted by the Acute Physiology and Chronic Health Evaluation III was 9.2% for the cases compared with 10.3% for the controls (p = .7127). The observed hospital mortality rate was 20% in the cases vs. 7% in the control group (p = .0085). When adjusted for the severity of illness and invasive mechanical ventilation, nicotine replacement therapy was independently associated with increased mortality (odds ratio, 24.6; 95% confidence interval, 3.6-167.6; p = .0011). The mean (SD) 28-day intensive care unit-free days were 20.7 (10.5) in the case group compared with 23.4 (7.1) in the control group (p = .0488). Conclusions, Our study shows that nicotine replacement therapy is associated with increased hospital mortality in critically ill patients. However, because of the limitations of the study, a future study based on a better case-control design is warranted.