Is inhaled prophylactic heparin useful for prevention and management of pneumonia in ventilated ICU patients?

Bandeshe. J Crit Care 2016; 34:95-102. doi:10.1016/j.jcrc.2016.04.005

Clinical Question

• In adult mechanically ventilated patients, does inhaled heparin compared to inhaled saline or usual care reduce the incidence of ventilator acquired pneumonia (VAP)?

Design

• Feasibility phase 2b study
• Multi-centre
• Randomised, controlled trial
• Double-blind design (treating physicians and patients)
• concealed allocation by off-site computer randomisation
• Permuted block stratification according to study centre and surgical/non-surgical patient status
• Powered at 80% to detect reduction in VAP from expected baseline of 12% to 6%, with statistical significant set at 0.05, if 914 patients recruited
• Trial terminated due to futility after reviewing baseline VAP incidence and a revised power calculation
• Intention to treat analysis

Setting

• Number of centres and location not specified
• “…the majority of patients were recruited from a single center”
• April 2011 to December 2013

Population

• Inclusion: Adult patients over 18 years old; under 24 hours invasive mechanical ventilation; likely to require at least a further 48 hours of mechanical ventilation
• Exclusion: pregnancy; treatment limitations defined; contraindications to heparin; systemic anticoagulation administration; previous enrolment
  • Routine heparin for thromboembolism prevention and renal replacement therapy was allowed
• 2103 screened; 214 randomised; none lost to follow-up
• Baseline characteristics
  • Mean age: 56 years
  • Mean APACHE II score: 18.9
  • Mean SOFA score: 6
  • Gender: 66% male
  • Pneumonia diagnosis on admission: 42%
  • Non-surgical admission: 64%

Intervention

• Heparin group
  • Nebulised unfractionated heparin 5000 units
  • Made up to 2 ml with 0.9% saline
  • administered every 6 hours

Control

• Saline group
  • 0.9% sodium chloride 2 ml
  • Administered every 6 hours
• Usual Care group
  • No nebulised heparin or saline
  • Allocation to this group was not ‘blinded’ to the clinical staff

Management common to all groups

• Study drug administered regularly until no longer mechanically ventilated for more than 48 hours or discharge from ICU
• If re-intubated for further mechanical ventilation, the study drug was administered according to the patient’s previous allocation
• Humidification and all non-saline/heparin nebulised therapies were allowed
• Pneumonia was treated according to standard recommendations, with at least 5 days antibiotics

Outcome

• Primary outcome: No difference in the incidence of VAP
  • Clinical diagnosis: heparin group 28% vs saline group 24% vs usual care group 26%; P=0.85
  • Klompas criteria: heparin group 7% vs saline group 6% vs usual care group 7%; P=1.0
• Secondary outcome: No significant differences found (heparin vs saline vs usual care)
  • Time to VAP: 7 days vs 7.5 days vs 8 days; P=0.35
  • Ventilator associated complications: 29% vs 15% vs 18%; P=0.59
  • Adverse events: 7% vs 6% vs 1%; P=0.27

Authors’ Conclusions

• Nebulised 5000 units of unfractionated heparin given four times a day is not recommended for the prophylaxis of ventilator associated pneumonia.

Strengths

• Good study methodology so no significant biases likely and probably good accuracy of results (internal validity)
  • Appropriate randomisation method and concealment
  • Blinded treatment and additional ‘usual care’ group to remove possible influence of nebulised saline
  • No patients lost to follow-up
  • Objective definition for primary outcome
• Broad inclusion criteria so good external validity and generalisability

Weaknesses

• Under-powered to firmly conclude that heparin has no effect
  • Point effect of Absolute Risk Reduction (ARR) for heparin group vs all control group is 0.75% (95% CI -6.41% to 7.91%)
  • Authors point out that, if this point effect is accurate and baseline is 6%, a trial with 22,000 patients is required to provide sufficient power for firm conclusion!
  • Although probably clinically negligible, this IPHIVAP trial cannot rule out heparin being beneficial (up to ARR 7.9%) or harmful (up to ARI 6.4%)
• With 90% excluded after screening and majority recruited from single centre, some may argue that results are not generalisable

The Bottom Line

• Nebulised heparin is probably ineffective, or the effect is clinically negligible, therefore I shall not be administering this to my patients to prevent ventilator associated pneumonia

External Links

• [article] Is inhaled prophylactic heparin useful for prevention and management of pneumonia in ventilated ICU patients?
• [further reading] What is the evidence for the use of nebulised heparin in cystic fibrosis
• [further reading] Nebulized anticoagulants for acute lung injury – a systematic review of preclinical and clinical investigations
• [further reading] Rapid and Reproducible Surveillance for Ventilator-Associated Pneumonia by Klompas

Metadata

Summary author: Duncan Chambler
Summary date: 9 February 2017
Peer-review editor: Steve Mathieu