From Wiki Journal Club
Jump to navigation Jump to search
Chastre J. "Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: a randomized trial". Journal of the American Medical Association. 2003. 290(19):2588-98.
PubMedFull textPDF

Clinical Question

In patients with ventilator-associated pneumonia (VAP), what is the optimal duration of antibiotic therapy in reducing mortality and microbiology recurrence?

Bottom Line

PneumA demonstrated equivalent mortality among patients with ventilator-associated pneumonia (VAP) treated with 8 days compared to 15 days of antibiotics, except when Pseudomonas aeruginosa was suspected based on gram stain.

Major Points

Few studies had addressed the optimal duration of antibiotic therapy in VAP, and consequently many providers routinely recommended prolonged 14- or 21-day antibiotic courses for such patients. However, heightened awareness of the association between prolonged antibiotics and multidrug-resistant infections raised concerns that perhaps this practice should be reconsidered. The PneumA trial compared two antibiotic durations for VAP, and demonstrated similar 28-day mortality among patients treated with 8 days compared to 15 days. A subgroup analysis suggested that a 15-day course may be superior in cases of VAP due to lactose nonfermenting GNR (eg, Pseudomonas aeruginosa).


  • Multicenter, double-blinded (until day 8), parallel-group, noninferiority, randomized, controlled trial
  • N=401:
    • Eight days (n=197)
    • Fifteen days (n=204)
  • Setting: 51 ICUs in France
  • Enrollment: May 1999 to June 2002
  • Analysis: Intention-to-treat


Inclusion Criteria

  • Mechanical ventilation for >48 hours
  • Age ≥18 years
  • Clinical suspicion of VAP
    • ARDS or new/persistent CXR infiltrate
    • One of the following:
      • Purulent tracheal secretion
      • Fever ≥38.3
      • WBC ≥10,000
  • Positive cultures from one of:
    • Fiberoptic bronchoscopy
    • Broncheoalveolar lavage fluid (≥104 cfu/ml)
    • Protected specimen brush or catheter (≥103 cfu/ml)
  • Appropriate empiric antibiotics within 24 hours

Exclusion Criteria

  • Pregnancy
  • Enrollment in another trial
  • SAPS II score >65 (indicating poor survival)
  • Neutropenia (WBC <1,000 or ANC <500)
  • AIDS
  • Receipt of immunosuppressants or long-term corticosteroids (≥0.5 mg/kg/day for >1 month)
  • Concomitant extrapulmonary infection between study days 1-3 requiring >8 days of antibiotics
  • Treating physician unwilling to use full life support
  • Onset of pneumonia within 5 days of initiating mechanical ventilation if no antibiotics received within prior 15 days


  • Randomly assigned to 8-day or 15-day duration group
  • Randomization occurred 3 days after bronchoscopy
  • Antimicrobial selection per treating physician
    • Protocol stipulated preference of aminoglycoside or fluoroquinolone and broad-spetrum beta-lactam
  • Antibiotics stopped at day 8 or 15 as determined by assigned group, although if antibiotics were required for treatment of ongoing infection that preceded VAP onset, these were continued


Comparisons are 8-day course vs. 15-day course.

Primary Outcomes

28-day mortality
18.8% vs. 17.2% (RR 1.1; P=0.41)
Microbiologic recurrence
28.9% vs. 26.0% (RR 1.1; P=0.16)
Mean number of antibiotic-free days
13.1 vs. 8.7 (RR 1.51; P=0.25)

Secondary Outcomes

No between-group difference reached statistical significance.

Ventilator-free days
8.7 vs. 9.1
Organ failure-free days
7.5 vs. 8.0
ICU days
30.0 vs. 27.5
Rate of unfavorable outcomes
46.2% vs. 43.6%
60-day mortality
25.4% vs. 27.9%
In-hospital mortality
32.0% vs. 29.9%

Subgroup Analysis

Recurrence among those with nonfermenting GNRs
40.6% vs. 25.4% (RD 15.2%; 90% CI 3.9% to 26.6%; P not specified)


Supported by the Délégation à la Recherche Clinique, Assistance Publique-Hopitaux de Paris.