Continued enteral nutrition until extubation compared with fasting before extubation

Continued enteral nutrition until extubation compared with fasting before extubation in patients in the intensive care unit: an open-label, cluster-randomised, parallel-group, non-inferiority trial 

Landais et al. The Lancet Respiratory Medicine 2023. 11:319-28. doi://10.1016/ S2213-2600(22)00413-1 

Clinical Question

  •  In patients ventilated for more than 48h in the Intensive Care Unit, is continued enteral nutrition compared with fasting prior to extubation, non-inferior with respect to extubation failure at 7 days?

Background

  • Extubation, the removal of an endotracheal tube, is a common procedure in intensive care units. Oropharyngeal dysfunction may be present in the hours post extubation, increasing the risk of aspiration and extubation failure
  • Extubation failure, requiring re-intubation within a week, occurs in 20% of patients, even when they meet established extubation criteria
  • There is a strong association between extubation failure and nosocomial pneumonia possibly due to the aspiration of gastric contents 
  • Clinicians favour gastric vacuity, and fasting is frequently imposed before extubation in ventilated patients, to reduce risk of aspiration secondary to gastric insufficiency
  • Fasting is common practice but is based on guidelines for scheduled surgery and is largely unevaluated in the intensive care unit patient
  • It is unknown if fasting improves patient outcomes and this unevaluated practice might delay extubation, increase workload, add challenges to glycaemic control, or reduce caloric intake. Patients may also be fasted but a clinical decision not to extubate may also be made

Design

  • Open-label, cluster-randomised, parallel-group 
  • Centres were randomly assigned (1:1) to continued enteral nutrition until extubation or 6h fasting with concomitant (continuous) gastric suctioning before extubation, applied to all patients within the unit 
  • Cluster randomisation, with no cross-over at a unit level simplifies the implementation of the strategy 
  • A statistician (not involved in recruitment) prepared the computer-generated randomisation list  
  • Randomisation stratified according to whether hospital was university affiliated and based on predominant case-mix
  • Standard care, led by the clinical teams, was present in both arms of study 
  • A study protocol was published which outlined methods and a statistical analysis plan  
  • The sample size calculation was based on an expected incidence of extubation failure of 16% in the control group, with a non-inferiority margin of 10%, a one-sided type 1 error of 2.5% and a power of 80%, and an interclass correlation coefficient of 0.03, 1100 participants were needed 
  • The analysis included the Intention-to-Treat group (those who were included regardless of protocol deviations) and the Per-Protocol Population (those who strictly adhered to the protocol) 
  • Nosocomial pneumonia (secondary outcome) was diagnosed by a centralised masked (blinded) committee 
  • Ethical approval was gained from the Comite de Protection des Personnes Sud Mediterranee 2017
  • Funded by the French Ministry of Health who were not involved with the design, data collection, data interpretation, or writing of the study 
  • The trial was registered with ClinicalTrials.gov, NCT03335345

Setting

  • 22 intensive care units in France
  • Data collected between April 2018 to October 2019

Population

  • Inclusion
    • Patients aged 18 years or older  
    • Received invasive mechanical ventilation for at least 48h  
    • Received prepyloric enteral nutrition for at least 24 h at the time of extubation decision 
  • Exclusion 
    • Pregnancy or breastfeeding 
    • Do-not-reintubate order 
    • Guardianship or trusteeship 
    • Absence of Social Security coverage 
    • Tracheostomy 
    • Post-pyloric enteral nutrition (jejunal tube) 
    • Previous inclusion in the trial
  • 7056 participants received enteral nutrition and mechanical ventilation, and were admitted to the intensive care units
  • 4198 were assessed for eligibility (2858 were excluded)
  • 1130 patients were enrolled (3068 were excluded from randomisation, leaving 617 in the continued enteral feeding arm and 513 in the fasting group)
  • According to French law, consent was obtained from the participant, next of kin, or surrogate decision maker. In all cases, the participants were informed about the study
  • Comparing baseline characteristics of continuous enteral feeding vs fasted group 
    • Age: 62 vs 64 years  
    • Male: 68% vs 63% 
    • SAPS II: 51 vs 50 
    • Types of admissions 
      • Unscheduled surgical: 11% vs 12% 
      • Scheduled surgical: 2% vs 6% 
      • Medical: 87% vs 82% 
    • Cough evaluation  
      • Ineffective: 2% vs 8% 
      • Weak 17% vs 26% 
      • Effective 68% vs 58% 
      • Very effective 14% vs 9% 
    • Administration of steroids before extubation: 17% vs 48% 
      • For prevention of laryngeal oedema: 3% vs 31% 
      • Other indication: 14% vs 17% 
  • Tracheal secretion burden was comparable between groups
  • The median duration of mechanical ventilation prior to extubation was similar in both groups: 8.4 days vs 8.3 days

Intervention

  • Continued enteral nutrition until extubation
    • No change to the enteral nutrition rate 
    • No suctioning of gastric tube prior, during or after extubation 
    • Median volume of enteral intake in 6h prior to extubation was 375ml

Control

  • Discontinuation of enteral feeding 6h before extubation 
    • Maximal gastric vacuity was implemented comprising of interruption of enteral feeding and concomitant continuous suctioning via the gastric tube when feasible
    • Median volume of gastric contents suctioned 6h prior to aspiration was 20ml (IQR 2-200ml)

Management common to both groups

  • Left to the discretion of the attending physician:
    • The decision of reintubation
    • Aspiration prevention measures, such as head-of-bed elevation, as well as the nutrition strategy before study inclusion
    • Post-extubation: the choice to maintain the gastric tube, the conduct of nutrition via the gastric tube, and the parenteral or oral route
  • In the case of reintubation and subsequent extubation, patients were managed according to the initial allocation group 

Outcome

  • Primary outcome: Extubation failure (composite criteria of reintubation or death) within 7 days, comparing continuous vs fasting group – no significant difference
    • In the Intention-to-Treat population: 106/617 (17.2%) vs 90/513 (17.5%)
      • Absolute difference -0.4% (95% CI: -5.2% to 4.5%)
    • Per-protocol population: 101/595 (17.0%) vs 74/413 (17.9%)
    • Post-hoc sensitivity analysis assessing the absolute difference between the groups (adjusted for potential confounders) confirmed the non-inferiority results
  • Secondary outcomes:
    • Comparing continuous vs fasting groups
      • No significant difference in
        • Pneumonia within 14 days of extubation
          • 1.6% vs 2.5% (rate ratio 0.77, 95% CI 0.22-2.69)
        • Caloric intake on the day of extubation
          • 880kcal vs 700kcal
      • Significantly reduced in continuous group
        • Hyperglycaemia on the day before extubation
          • 18.3% vs 26.5%
          • No significant difference in hyperglycaemia episodes at other timepoints
        • Hypoglycaemia on the day of extubation
          • 0.3% vs 3.1%
          • No difference in hypoglycaemia episodes at other timepoints
        • Duration between first spontaneous breathing trial and extubation
          • 2.0h vs 17.6h
        • Duration between first spontaneous breathing trial and ICU discharge
          • 4.0 days [IQR 2.0-8.0] vs 6.6 days [3.0-11.1], HR 1.45, 95% CI 1.19-1.77)
        • ICU mortality
          • 3.9% vs 6.8% (RR0.56, 95% CI 0.32-0.99)
      • Significantly increased in continuous group
        • Extubation failure at 48 hours
          • 12.5% vs 8.6%
        • Extubation failure at 72 hours
          • 14.7% vs 10.5%

Authors’ Conclusions

  • The authors conclude that continued enteral feeding in critically ill patients until extubation was non-inferior to 6h fasting maximum gastric vacuity strategy in terms of re-intubation rates within 7 days

Strengths

  • The primary outcome was appropriate, being a composite outcome of death or reintubation, and accounted for the minority of patients who might die without being reintubated
  • Multi-centre
  • Cluster randomisation avoided cross-contamination of practice and was an appropriate level to randomise at (unit level rather than individual patient level)
  • Post-hoc sensitivity analysis assessing the absolute difference between the groups (adjusted for potential confounders) confirmed the non-inferiority results
  • Blinding of outcome assessors of nosocomial pneumonia

Weaknesses

  • Open-label and therefore treating clinicians were not blinded to treatment arm 
  • Characteristic differences exist between the continuous feeding and the fasted group:
    • Patients in the fasting group were older and were more likely to have an ineffective cough which may have influenced their extubation failure
    • Patients in the intervention group were sicker as evidenced by SAPS II scores, were less likely to have passed their spontaneous breathing trial, and had fewer steroids to treat laryngeal oedema
  • The trial was a cluster randomised trial, but there was no crossover with each site only randomised to either the intervention or the control arm. The decision of extubation, re-intubation and post-extubation nutritional strategy was left to the discretion of the attending physician.  This introduces the potential for variation in patient care and may have subsequently affected the data. The very large difference in pre-extubation steroid use suggests that there are differences in the management strategies of the centres randomised to the intervention vs the control arm
  • The application of continuing feed in major abdominal surgery remains uncertain
  • Maximum gastric vacuity practices may not mirror standard practice in other intensive care units  
  • It would be helpful to see the mortality breakdown of patients given that all-cause mortality was calculated 
  • The exclusion of patients with no social security coverage is not explained. This exclusion may introduce accessibility and equity issues in research participation. Furthermore, limiting participants based on social security coverage could reduce the study’s applicability to broader populations 
  • The cough evaluation process is not explicit, and it is unclear if this evaluation used a tool and was consistent across all participating sites
  • A large number of patients were excluded, thus limiting external validity

The Bottom Line

  • Continuous enteral nutrition prior to extubation was non-inferior when compared to 6h fasting pre-extubation
  • Based on this study,
    • We would advocate continuously feeding our patients that match the demographics of this study prior to extubation
    • We would be cautious in patients with ileus, high gastric aspirates, prolonged ventilation, or an increased risk of oropharyngeal dysfunction following extubation

External Links

 

Metadata

Summary author: Jody Ede, Andrew Ellis, Fran Piccaver
Summary date: 16/12/24
Peer-review editor: Dave Slessor

Picture by: Ede

 

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.