Effect of Fluid Bolus Administration on Cardiovascular Collapse Among Critically Ill Patients Undergoing Tracheal Intubation

@DerekRussellMD. JAMA; 328(3):270-279. doi:10.1001/jama.2022.9792

Clinical Question

• In critically ill adult patients undergoing tracheal intubation and positive pressure ventilation, does intravenous infusion of a 500ml fluid bolus vs no fluid bolus, decrease the incidence of cardiovascular collapse during, or shortly after the procedure?

Background

• Tracheal intubation in the intensive care unit is a common intervention that entails significant risk
• Haemodynamic instability, in the form of hypotension, increased vasopressor requirement or cardiac arrest, is a common complication, occurring in around 25% of intubations in the critical care setting, compared with only 4% in the theatre setting
• There are a number of potential causes, including loss of sympathetic tone, increased intrathoracic pressure and the use of anaesthetic induction agents
• A simple potential intervention would be a fluid bolus to increase intravascular filling
•   The only prior study on this intervention was PREPARE, which was stopped early due to futility
• A post hoc subgroup analysis of PREPARE found a potential benefit in patients undergoing positive pressure ventilation during the intubation process

Design

• Multi-centre, randomised, unblinded trial
• 1:1 randomisation to intervention and control in randomly permuted block sizes of 2, 4 and 6; stratified by trial site
• Informed consent was obtained after the intervention
• Sample size calculation: 1065 patients would give 80% power to detect a between group relative difference of 35%
• Primary outcome was cardiovascular collapse, defined as one or more of:
  • New or increased receipt of vasopressors between induction and 2 minutes after tracheal intubation
  • Systolic blood pressure of less than 65 mmHg between induction and 2 minutes after tracheal intubation
  • Cardiac arrest between induction and 1 hour after tracheal intubation
  • Death between induction and 1 hour after tracheal intubation
• Sole pre-specified secondary outcome was in-hospital mortality by day 28

Setting

• 11 intensive care units in the US
• Enrollment occurred between February 2019 and May 2021
• Enrollment paused between February 2020 and August 2020 due to Covid-19 pandemic

Population

• Inclusion:
  • Adult patients over 18 years old
  • Undergoing tracheal intubation, with a plan to use
    • Anaesthetic induction agents
    • Positive pressure ventilation during induction and prior to direct laryngoscopy
• Exclusion:
  • Immediate need for intubation
    • These cases were deemed too urgent to allow time for randomisation
  • If the intubating clinician felt the intervention was either indicated or contra-indicated
  • Pregnancy
  • Incarceration
• 1576 patients screened
  • 1065 randomised
    • 538 fluid bolus
    • 527 no fluid bolus
• Comparing baseline characteristics of fluid bolus vs. no fluid bolus group
  • Median age (years): 61 vs 62
  • Female Sex (%): 41 vs 43
  • BMI (kg/m²): 28 vs 28
  • Comorbidities (%):
    • Diabetes: 26 vs 29
    • Hypertension: 34 vs 34
    • COPD: 19 vs 17
    • Congestive cardiac failure: 15 vs 14
    • Renal failure: 4 vs 6
  • Indications for intubation (%)
    • Hypoxia: 41 vs 43
    • Hypercarbia: 7 vs 8
    • Hypoxia and hypercarbia: 11 vs 11
    • Low GCS: 20 vs 20
  • Use of vasopressors (%): 20.0 vs 19.4
  • Receiving IV fluids at enrollment (%): 10.2 vs 9.9

Intervention

• Fluid bolus group
  • 500mls of isotonic crystalloid solution, intravenously
  • Aim to infuse as much as possible of the 500mls before induction of anaesthesia, without delaying tracheal intubation
  • Complete infusion of any remaining fluid after induction, during tracheal intubation

Control

• No fluid bolus group
  • Initiation of fluid bolus not permitted, except
    • As treatment for hypotension
    • If operator determined that fluid bolus was necessary for patient safety

Management common to both groups

• All other aspects of care at clinicians’ discretion
• Specifically
  • Ongoing IV fluid infusions could be continued
  • Vasopressors to treat hypotension
  • Choice of induction agent

Outcome

• Primary outcome: incidence of cardiovascular collapse – no significant difference
  • 21.0% in fluid bolus group vs 18.2% in no fluid bolus group
    • Absolute difference 2.8%, 95% CI -2.2% to 7,7%
• Secondary outcome: in-hospital mortality at 28 days – no significant difference
  • 40.5% in fluid bolus group vs 42.3% in no fluid bolus group
• No difference in any of the prespecified exploratory outcomes

Authors’ Conclusions

• In critically ill adults undergoing tracheal intubation with positive pressure ventilation, a fluid bolus during induction of anaesthesia did not decrease the incidence of cardiovascular collapse, compared with no fluid bolus

Strengths

• Pragmatic, randomised design
• Clinically relevant question with an easily implemented intervention
• Conclusions seem to be broadly generalisable, with patient characteristics that seem very similar to patients on my unit
  • Co-morbidities
  • Acute presentation
  • Indications for intubation
• Use of composite outcome increases probability of finding a clinically important difference
• Very impressive implementation of interventions, and separation between the two groups
  • 99.4% of fluid bolus group received bolus, compared with 1.1% in no fluid bolus group
  • Median of 500mls of fluid administered in fluid bolus group at 2 minutes after intubation, compared with 0mls in no fluid bolus group
  • Lowest volume administered was 300mls
• Only a small variation in fluid management between the the two groups following induction
  • A large variation could have introduced bias

Weaknesses

• One could argue that fluid administration does not address the main causes of cardiovascular collapse during intubation
  • i.e. the hypothesis lacks face validity
• The selection of induction agents is not representative of the agents used in most units outside the US
• Issues with using a composite outcome:
  • The components of the composite outcome are not all equally important
    • for example, cardiac arrest is devastating compared to an episode of hypotension
  • The components of the composite outcome are not all patient-centred
    • patients unlikely to be effected by an episode of hypotension or increased dose of vasopressor
  • Some elements of the composite outcome susceptible to bias
    • Operators’ use of vasopressors may be affected by the trial intervention
    • Increased use of vasopressors may not reflect genuine cardiovascular collapse or hypotension
• Unblinded trial – could it have been possible, albeit more difficult, to blind clinicians to the intervention?
• Large proportion of patients (15%) were excluded due to urgency – potential cause of bias

The Bottom Line

• The routine administration of isotonic crystolloid solution during induction of anaesthesia and intubation does not reduce the incidence of cardiovascular collapse
• I will continue to assess my patients’ fluid status on a case by case basis, and will only administer a fluid bolus in patients who display convincing evidence of hypovolaemia
• I will continue to proactively administer a dose of vasopressor during induction in patients who I consider at particular risk of cardiovascular collapse

External Links

• Effect of Fluid Bolus Administration on Cardiovascular Collapse Among Critically Ill Patients Undergoing Tracheal Intubation
• Trial Results Presentation

Metadata

Summary author: Omar Jundi – @bi0marker
Summary date: 16/12/2022
Peer-review editor: David Slessor

Picture by: author/site