Effect of Tele-ICU on Clinical Outcomes of Critically Ill Patients

Importance Despite its implementation in several countries, there has not been a randomized clinical trial to assess whether telemedicine in intensive care units (ICUs) could improve clinical outcomes of critically ill patients. Objective To determine whether an intervention comprising daily multidisciplinary rounds and monthly audit and feedback meetings performed by a remote board-certified intensivist reduces ICU length of stay (LOS) compared with usual care. Design, Setting, and Participants A parallel cluster randomized clinical trial with a baseline period in 30 general ICUs in Brazil in which daily multidisciplinary rounds performed by board-certified intensivists were not routinely available. All consecutive adult patients (aged ≥18 years) admitted to the participating ICUs, excluding those admitted due to justice-related issues, were enrolled between June 1, 2019, and April 7, 2021, with last follow-up on July 6, 2021. Intervention Remote daily multidisciplinary rounds led by a board-certified intensivist through telemedicine, monthly audit and feedback meetings for discussion of ICU performance indicators, and provision of evidence-based clinical protocols. Main Outcomes and Measures The primary outcome was ICU LOS at the patient level. Secondary outcomes included ICU efficiency, in-hospital mortality, incidence of central line–associated bloodstream infections, ventilator-associated events, catheter-associated urinary tract infections, ventilator-free days at 28 days, patient-days receiving oral or enteral feeding, patient-days under light sedation, and rate of patients with oxygen saturation values under that of normoxemia, assessed using generalized linear mixed models. Results Among 17 024 patients (1794 in the baseline period and 15 230 in the intervention period), the mean (SD) age was 61 (18) years, 44.7% were female, the median (IQR) Sequential Organ Failure Assessment score was 6 (2-9), and 45.5% were invasively mechanically ventilated at admission. The median (IQR) time under intervention was 20 (16-21) months. Mean (SD) ICU LOS, adjusted for baseline assessment, did not differ significantly between the tele–critical care and usual care groups (8.1 [10.0] and 7.1 [9.0] days; percentage change, 8.2% [95% CI, −5.4% to 23.8%]; P = .24). Results were similar in sensitivity analyses and prespecified subgroups. There were no statistically significant differences in any other secondary or exploratory outcomes. Conclusions and Relevance Daily multidisciplinary rounds conducted by a board-certified intensivist through telemedicine did not reduce ICU LOS in critically ill adult patients. Trial Registration ClinicalTrials.gov Identifier: NCT03920501