PROPHY-VAP: Ceftriaxone to prevent early ventilator-associated pneumonia

8 February 2024 andrew.achilleos@nhs.net Intensive Care Medicine, Summaries Leave a comment

Ceftriaxone to prevent early ventilator-associated pneumonia in patients with acute brain injury: a multicentre, randomised, double-blind, placebo-controlled, assessor-masked superiority trial

C Dahyot-Fizelier @ClaireDahyot Lancet 2024 Jan 19:S2213-2600(23)00471-X. doi: 10.1016/S2213-2600(23)00471-X.

Clinical Question

In patients with severe brain injury who require mechanical ventilation, does an early, single dose of ceftriaxone compared with placebo reduce the incidence of early VAP?

Background

Ventilator-associated pneumonia (VAP) is common in critically unwell patients and is associated with increased antibiotic prescription, as well as negative patient centered outcomes including prolonged duration of mechanical ventilation and increased mortality
Acute brain injury further increases the risk of early VAP, likely due to aspiration of microorganisms prior to intubation, so this may be a particularly useful cohort to target with preventative strategies
VAP prevention with selective digestive tract decontamination (SDD) has had favourable results in numerous clinical trials, and a secondary analysis of a recent clinical trial suggested a particular benefit for brain injured patients. However, there is reluctance to embrace this treatment due to concerns that it could promote antibiotic-resistant pathogens
It is hypothesised that a single prophylactic dose of ceftriaxone in brain injured patients may reduce the incidence of VAP, and paradoxically reduce the risk of antibiotic resistance through reduced overall antibiotic prescription

Design

Multicentre, randomised, double-blind, placebo-controlled superiority trial
Conducted in nine ICUs of eight French university hospitals
Eligible patients were randomly assigned (1:1) by a web-based randomisation system, stratified by centre and severity of unconsciousness at the time of inclusion using a computer-generated sequence with variable block sizes
Patients, health-care providers, assessors, and the study statistician were masked to the allocation group
The sample size was based on a mean incidence of early VAP of 30% in the control group, and the hypothesis that it could be reduced by half in the intervention group (15%)
On this basis, the investigators calculated that a sample size of 354 patients would provide 90% power at an α of 0.05
Intention-to-treat analysis
The trial protocol, including the statistical analysis plan, was published a priori

Setting

345 patients recruited between October 2015 and May 2020, of whom 171 were assigned to receive ceftriaxone and 174 to placebo
15 patients did not receive the allocated intervention; 6 from the ceftriaxone group and 9 from the placebo group, and 11 patients withdrew consent; 3 from the ceftriaxone group and 8 from the placebo group
319 patients were included in the analysis, of whom 162 were assigned to receive ceftriaxone and 157 to placebo

Population

Inclusion: Brain injured patients (head trauma, stroke, or subarachnoid haemorrhage), aged ≥18 years with a GCS ≤12 who had been intubated for less than 12 hours, with an expected duration of MV more than 48 hours
Exclusions: The main exclusions were as follows:
- Intubation ≥48 hours after admission
- ETT with subglottic secretion drainage
- Coma due to tumour, infectious disease or cardiac arrest
- Hospitalisation within the last month
- Beta-lactam allergy
- In receipt of antibiotics at time of admission
Baseline demographics (ceftriaxone vs placebo):
- Age: 56 vs 57 years
- Male sex: 49% vs 55%
- BMI: 26 vs 26 kg/m2
- Medical History:
  - Smoking: 21% vs 22%
  - Alcoholism: 17% vs 15%
  - Diabetes: 4% vs 5%
  - Chronic lung disease: 3% vs 1%
- Main severe brain injury:
  - Ischaemic stroke: 10% vs 8%
  - Haemorrhagic stroke: 22% vs 21%
  - Subarachnoid haemorrhage: 38% vs 45%
  - Brain trauma: 30% vs 26%
- GCS:
  - 3: 17% vs 13%
  - 4-8: 60% vs 65%
  - 9-12: 23% vs 22%
- SAPS score: 47 vs 48
- Time from tracheal intubation: 7 hours vs 7 hours
- Mouth care with chlorhexidine: 4% vs 4%

Intervention

Ceftriaxone
- Single administration of 2g intravenous ceftriaxone

Control

Placebo
- Single administration of intravenous saline

Management common to both groups

Opaque syringes and infusion lines as ceftriaxone in solution is slightly coloured
VAP prevention measures were routinely implemented before trial commencement including hand washing, 30° head-of-bed elevation, monitoring of ETT cuff pressure, mouth care and ulcer prevention
None of the participating centres performed SDD
VAP was declared according to American Thoracic Society criteria, which rely on clinical, radiological and microbiological criteria occurring at least 48 h after the start of mechanical ventilation
A central adjudication committee composed of two senior intensivists masked to study group assignment, reviewed all declared cases
Patients who developed VAP (or other infection) received curative antibiotic therapy according to local protocols, based on national guidelines

Outcome

Primary efficacy endpoint: The incidence of early VAP (from 2nd – 7th day) was significantly reduced in the ceftriaxone group
- 14% vs 32% (HR 0.60; 95% CI 0.38 to 0.95)
Secondary efficacy endpoints:
- There was a significant increase in ventilator and antibiotic-free days, and a significant reduction in mortality and the incidence of all VAP in the ceftriaxone group at day 28
  - All VAP: 20% vs 36%
  - Ventilator-free days: 9 vs 5
  - Antibiotic-free days: 21 vs 15
  - Mortality: 15% vs 25%

Safety outcomes:
- There were 90 serious adverse events, which were all attributed to the initial disease and not to trial intervention
- 3 patients developed Clostridium difficile infection; 1 in the ceftriaxone and 2 in the placebo group
- There was 1 MRSA strain and 1 ESBL-producing Enterobacteriaceae strain in the placebo group

Authors’ Conclusions

“In patients with brain injury who required ICU admission and mechanical ventilation, we found that early administration of a single ceftriaxone dose decreased the risk of VAP, exposure to ventilation, exposure to antibiotics, prolonged ICU and hospital stay, and mortality, with no safety concerns.”

Strengths

Multi-centre, randomised, controlled trial
Blinded nature of the study and independent verification of VAP diagnosis reduced the potential for performance and reporting bias
Central randomisation with variable blocks ensured allocation concealment
Baseline demographics were well matched between groups
Most patients received treatment according to their randomised group and there was minimal attrition
The choice of ceftriaxone as the prophylactic antibiotic was sensible given its microbiological cover and long half-life
Intention to treat analysis preserved the randomisation schedule
Pragmatic inclusion and exclusion criteria captured the group of interest and enhanced external validity
The authors hypothesis that overall antibiotic prescription would decrease in the intervention group was supported
This may be the most compelling form of VAP prevention – it has considerably better fidelity than SDD, which means that clinicians are more likely to use it, and unlike AMIKINHAL, had important patient-centered secondary outcome benefits, in particular mortality

Limitations

Other forms of VAP prevention were not monitored, so it is not possible to know to what extent this could have impacted results
The blinding procedure was not performed remotely, but by nurses from neighbouring units, which could have caused unmasking of the treatment. However, these nurses were not involved in direct patient care and opaque syringes and infusion lines were used
The adjudication committee that reviewed suspected cases of VAP was not independent
The choice of primary outcome could be considered controversial. VAP detection is notoriously inaccurate, and there are numerous possibilities for confounding, such as fever which is common in brain injured patients. The diagnosis of VAP also required required positive microbiology which may have been more difficult to detect following the administration of antibiotics
The 10% difference in mortality is unlikely to be caused by the intervention alone. Therefore there may be some confounding variables that have impacted the results reported
Whilst the secondary outcome benefits are compelling, the study was not powered for these. Secondary outcomes are generally considered to be hypothesis generating. Those with enthusiastic priors may point to the overall weight of evidence in conjunction with other recent VAP prevention studies
External validity was somewhat limited as the study was conducted only in France
It may be hard to conclude that there was no increased risk of antibiotic resistant pathogens, as the study was underpowered for this outcome and there were very few events. However, the reduction in overall antibiotic prescription may be considered a reasonable surrogate and SDD studies with more longitudinal follow up have generally not supported this concern

The Bottom Line

Whilst this is a historically divisive topic, it is becoming harder to argue against the consistent signal that a short course of prophylactic antibiotics among patients who are intubated due to acute brain injury is beneficial and safe
This well executed study provides evidence that an early, single dose of ceftriaxone prevents early VAP in patients with severe brain injury, as well as decreasing antibiotic and ventilation exposure, and mortality without safety concerns, and I will changing my practice accordingly

External Links

Metadata

Summary author: Andrew Achilleos
Summary date: 27th January 2024
Peer-review editor: David Slessor

Image by: CDC on Unsplash