Diagnostic Methods Equivalent for Mechanical Ventilation-Related Pneumonia

KINGSTON, Ontario, Dec. 20 -- For critically ill patients on mechanical ventilation, bronchoalveolar lavage and endotracheal aspiration appear to be equally effective when diagnosing pneumonia, said researchers here.

Among 740 patients in 28 North American intensive care units, there was no difference in clinical outcomes or antibiotic use associated with the diagnostic methods, reported Daren Heyland, M.D., of Kingston General Hospital and Queens University here, and colleagues in the Dec. 21 issue of the New England Journal of Medicine.

Many clinicians see bronchoalveolar lavage as more accurate, which would give them the confidence to stop unnecessary antibiotics or use narrow spectrum, targeted antibiotics that may help reduce antibiotic resistance, said Marin H. Kollef, M.D., of Washington University and Barnes-Jewish Hospital in St. Louis, in an accompanying editorial.

"The culture of bronchoalveolar-lavage fluid is more likely to result in modification of prescribed broad-spectrum regimens than is the culture of an endotracheal aspirate," he wrote.

However, the researchers found that among patients without antibiotic-resistant infections, the two diagnostic methods resulted in similar rates of targeted antibiotic therapy (74.2% and 74.6%, respectively, P=0.90).

The researchers said they expected to find better outcomes with bronchoalveolar-lavage but instead found the differences were insignificant. For bronchoalveolar-lavage versus endotracheal-aspiration, they reported:

• Similar 28-day mortality rates, the primary endpoint, between groups (18.9% versus 18.4%, P=0.94).
• No significant difference in days alive without antibiotics (10.47.5 versus 10.67.9, P=0.86).
• No significant difference in the time to intensive care unit discharge (12.3 days versus 12.2 days, P=0.22) or hospital discharge (40.2 days, versus 47.0 days, P=0.13).

Because about 20% of ICU patients on mechanical ventilation develop pneumonia, several studies have looked at the best diagnostic approach but with conflicting results. So the researchers used a larger trial comparing initial antimicrobial strategies to attempt to answer the question in a select group of patients.

They randomized immunocompetent patients with suspected ventilator-associated pneumonia after at least four days in the ICU to undergo bronchoalveolar lavage with quantitative culture of the fluid recovered or endotracheal aspiration with nonquantitative culture of the aspirate. After these tests, patients were randomized to initial antibiotic treatment with the 1 g of Merrem (meropenem) every eight hours plus 400 mg of Cipro (ciprofloxacin) every 12 hours or Merrem alone.

After culture results were reported, the treating physicians were to use a single antibiotic with the narrowest spectrum targeted to the bacteria found in order to discontinue or reduce the dose or number of antibiotics. Antibiotics were discontinued for negative cultures except at the physician's discretion among patients with a high pretest likelihood of ventilator-associated pneumonia.

The study excluded patients known to be infected with pseudomonas species or methicillin-resistant Staphylococcus aureus (MRSA), allergic to common antibiotic drugs, recent recipients of the study drugs, or expected to die or have a prolonged intensive care unit stay.

Results and diagnoses were centrally adjudicated after patient discharge or death.

The researchers reported no significant differences in baseline characteristics between groups including antibiotic use before enrollment. Positive cultures were found for 59.7% of the bronchoalveolar-lavage group and 51.9% of the endotracheal-aspiration group (P=0.03), but a similar percentage in both groups were found not to have ventilator-associated pneumonia (13.7% and 17.1%, respectively, P=0.19).

The adjusted relative risk of 28-day mortality was not significantly different between bronchoalveolar-lavage and endotracheal-aspiration (1.01, 95% CI 0.75 to 1.37, P=0.94). Nor were there any significant differences for any subgroup.

The investigators also found no difference in treatment effect according to which initial antibiotic therapy was used or for the antibiotic therapies according to which diagnostic test was used (P=0.37 for the interaction).

For bronchoalveolar-lavage and endotracheal-aspiration, the authors also reported no differences in secondary outcome endpoints. The respective findings were:

• Time to mechanical ventilation discontinuation after randomization was a median 8.9 days (95% CI 7.4 to 10.7), versus 8.8 days (95% CI 7.0 to 10.7, P=0.31).
• Days alive without antibiotics were 10.47.5 versus 10.67.9 (P=0.86).
• Maximum organ-dysfunction scores were 8.33.6 and 8.64.0 (P=0.26).

The lack of difference may be explained by a good antibiotic usage protocol, the researchers said. In the study, research personnel monitored patients and reminded clinicians to review culture results and adjust antibiotics as soon as the results were available.

However, since there was a very low prevalence of pseudomonas (6.4% overall) and methicillin-resistant S. aureus (1.6% overall), the findings may not be generalizable to settings with higher prevalence.

Dr. Kollef noted that at least 40% of the screened patients who were excluded had risk factors for colonization or infection with potentially antimicrobial-resistant bacteria.

"Unfortunately, these exclusions probably represent the majority of patients undergoing real-time evaluation for suspected ventilator-associated pneumonia," which diminishes the clinical utility of the results, he wrote.

Dr. Kollef concluded that despite these limitations, the study indicates that both strategies can be effective when used properly.

"Given the rapid emergence of antimicrobial resistance and the limited number of new antimicrobial agents, clinicians treating patients with suspected ventilator-associated pneumonia not only must prescribe appropriate initial antimicrobial regimens to optimize outcomes but also must minimize the development of resistance by rigorously using a de-escalation strategy," he wrote.

"When applied properly, bronchoalveolar lavage and endotracheal aspiration are tools that can facilitate de-escalation," he added.