番茄社区

SAN FRANCISCO -- Stopping enteral nutrition hours before an attempt at extubation didn't reduce risk of failure of extubation, a randomized trial showed.

Continued enteral nutrition until extubation carried a similar risk for the composite of reintubation or death within 7 days after extubation compared with a strategy of fasting for 6 hours and gastric suctioning before extubation.

The absolute difference of 0.4% between groups (17.2% vs 17.5%) met noninferiority criteria, reported Stephan Ehrmann, PhD, of the University of Tours in France, at the Society of Critical Care Medicine Critical Care Congress. The findings were published simultaneously in .

No subgroup failed to meet noninferiority between the two strategies, nor were there any differences in incidence of nosocomial pneumonia up to 14 days after extubation (1.6% nonfasting vs 2.5% fasting, rate ratio [RR] 0.77, 95% CI 0.22-2.69).

These unique, high-quality intervention trial findings "should prompt an immediate radical change in clinical practice," argued Elisabeth Lobmeyr, MD, of the Medical University of Vienna, and Karin Amrein, MD, MSc, of the Medical University of Graz in Austria, in an accompanying .

According to one from the U.K., 93% of intensive care units (ICUs) have patients fast before extubation, although with a wide range of durations from 1-2 hours to 8-12 hours.

For patients, eliminating the fasting improves their caloric intake, Lobmeyr and Amrein noted. "Patients receiving mechanical ventilation are often underfed, and interruptions are one of the main causes that energy prescriptions are not met."

In the study, calories actually delivered to the non-fasted group on the day of extubation dropped to about one-third of what they got the day prior (637 vs 1,775 kcal), whereas it dropped to about 20% in the fasting group (311 vs 1,505 kcal).

For hospital staff, it might mean less work for nurses, the editorialists emphasized: "Spontaneous breathing trials, preparation for extubation, and post-extubation phase are extremely busy periods for nurses. In times of post-pandemic staff shortages, any practice change resulting in less work for nursing staff is very welcome and should be embraced as soon as possible."

The pragmatic, open-label trial included 1,130 adults treated at 22 ICUs in France from April 2018 through October 2019, randomized by ICU as a cluster. Patients were included in the trial if they had received invasive mechanical ventilation for at least 48 hours in the ICU and received prepyloric enteral nutrition for at least 24 hours at the time of the decision to extubate.

More than 80% of the cohort were medical ICU patients, with an average of four comorbidities and a relatively low ICU mortality of 5% overall.

The 1,008 out of 1,130 patients enrolled who were in the per-protocol population had similar outcomes as the intention-to-treat population, with extubation failure in 17.0% of the continued enteral nutrition group versus 17.9% in the fasting group (absolute difference -0.9%, 95% CI -5.6 to 3.7).

Among the secondary endpoints, the non-fasting group had a significantly shorter median duration from the first successful spontaneous breathing trial to extubation (2.0 vs 17.6 hours) and from that trial to discharge alive from the ICU (4.0 vs 6.6 days; HR 1.45, 95% CI 1.19-1.77). Death in the ICU was significantly less common with continued enteral nutrition (3.9% vs 6.8%; RR 0.56, 95% CI 0.32-0.99).

Notably, while the presumed goal of fasting before extubation is to avoid aspiration, these events prompting reintubation were uncommon in both groups (8 vs 15 cases, respectively).

Clinicians are "probably right to perceive this risk," Ehrmann said at the session, citing studies showing that some 20% of patients have clinical dysphagia after extubation and that half of aspiration is silent.

Furthermore, along with being frequent, a retrospective study showed that significant dysphagia after extubation is associated with extubation failure, pneumonia, and mortality, he pointed out. "So that's not something to neglect."

However, fasting does not appear to be the proper response, he concluded, based on his group's findings. "The whole issue here is it's a very infrequent reason for reintubation, so it's not a good rationale."

His group acknowledged limitations of the study, including the open-label design (although nosocomial pneumonia diagnosis was adjudicated by a centralized masked committee) and cluster randomization without a crossover design. "The heterogeneity of units and enrolling physicians cannot be ignored," the editorialists wrote, although the multivariate-adjusted analysis affirmed the findings.