Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Prophylactic antibiotics reduce mortality in critically ill adults

ACP J Club. 1998 Sep-Oct;129:31. doi:10.7326/ACPJC-1998-129-2-031

Source Citation

D'Amico R, Pifferi S, Leonetti C, et al. Effectiveness of antibiotic prophylaxis in critically ill adult patients: systematic review of randomised controlled trials. BMJ. 1998 Apr 25;316:1275-85.



In critically ill cancer patients, does antibiotic prophylaxis reduce respiratory tract infections (RTIs) and overall mortality?

Data sources

Studies were identified by using MEDLINE (1984 to 1996) with the search terms intensive care units, critical care, antibiotics, respiratory tract infections, and SDD (selective decontamination of the digestive tract). Other sources were previous meta-analyses, conference proceedings and lists of investigators, and personal contacts.

Study selection

Published and unpublished randomized controlled trials were selected if they tested the effect of antibiotic prophylaxis in reducing RTIs and overall mortality in unselected critically ill adults. Eligible trials were classified as combined topical and systemic antibiotics compared with no treatment, or topical antibiotics alone compared with systemic antibiotics or placebo.

Data extraction

Data were extracted on sample size, treatment allocation, randomization and blinding methods, antibiotic type and dose, patients with ≥ 1 RTIs, deaths, and excluded patients. Separate meta-analyses were done by using aggregate (n = 33 studies) and individual patient data (n = 25 studies).

Main results

Meta-analyses of aggregate data from 30 trials (4898 patients) reporting on RTIs and 33 trials (5727 patients) reporting on mortality were done by using a fixed-effects model. Topical antibiotics combined with systemic antibiotics or used alone reduced RTIs { P < 0.001 for both}* (Table). Combined topical and systemic antibiotics reduced overall mortality { P < 0.001}* (Table), but topical antibiotics alone did not {24.2% vs 24.0%, P = 0.9}*. Meta-analysis of data on individual patients showed similar results.


Topical antibiotics combined with systemic antibiotics or used alone reduce respiratory tract infections in critically ill adult patients. Combined topical and systemic antibiotics also reduce overall mortality.

Source of funding: Hoechst Marion Roussel.

For correspondence: Dr. A. Liberati, Italian Cochrane Centre, Laboratory of Health Services Research, Mario Negri Institute, Via Eritrea 62, 20157 Milan, Italy. FAX 39-2-3560461.

* P values calculated from data in article.

Table. Combined topical and systemic antibiotics vs placebo for critically ill adults†

Outcomes to hospital discharge Weighted event rate RRR (95% CI) NNT (CI)
Combined Placebo
Respiratory infections 16.9% 34.1% 54% (48 to 60) 6 (5 to 7)
Mortality 24.7% 29.4% 18% (9 to 27) 22 (14 to 50)

†Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.


The median incidence of ventilator-associated lower RTI is > 10 episodes per 1000 ventilator days (1). Nosocomial lower RTI in critically ill patients is clearly an unsolved clinical problem. D'Amico and colleagues have shown, using meta-analysis, a reduction in the incidence of pneumonia and in mortality in critically ill patients, a testimony to both the individual studies and the power of meta-analysis. With the inclusion of disparate studies, definitions were broad; pneumonia and tracheobronchitis were considered together.

One concern addressed by the authors involved the selection of resistant flora. Use of topical prophylaxis has previously been shown to be associated with outbreaks of unusual flora (2); in critical care and other settings, antimicrobial prophylaxis has been associated with increases in resistant organisms (3, 4). Future trials may need to look specifically at these problems.

The analysis did not show that the benefit of combined prophylaxis was restricted to specific patient subgroups stratified by disease severity scores. However, studies based on specific preselected groups of patients were excluded from the analysis. Optimal use of antimicrobial prophylaxis strategies would logically be targeted at certain high-risk groups rather than being applied across the board because of antimicrobial resistance considerations, but the evidence to date does not support selective use.

David M. Miller, MD
Associates in Infectious and Tropical MedicinePittsburgh, Pennsylvania, USA


1. National Nosocomial Infections Surveillance System. Semiannual report. Summary of NNIS data. US Department of Health and Human Services and Centers for Disease Control. May 1996 [unpublished].

2. Brown RB, Phillips D, Barker MJ, et al. Outbreak of nosocomial Flavobacterium meningosepticum respiratory infections associated with use of aerosolized polymyxin B. Am J Infect Control. 1989;17:121-5.

3. Verwaest C, Verhaegen J, Ferdinande P, et al. Randomized, controlled trial of selective digestive decontamination in 600 mechanically ventilated patients in a multi-disciplinary intensive care unit. Crit Care Med. 1997; 25:63-71.

4. Fukatsu K, Saito H, Matsuda T, et al. Influences of type and duration of antimicrobial prophylaxis on an outbreak of methicillin-resistant Staphylococcus aureus and on the incidence of wound infection. Arch Surg. 1997; 132:1320-5.