Therapy Antibiotic prophylaxis of early onset pneumonia in critically ill comatose patients. A randomized study.
This review may be edited
Summary
Posted By:	Stuart Green
E-Mail:	stuart_green@health.qld.gov.au
Posted Date:	25/5/2005
Title:	Antibiotic prophylaxis of early onset pneumonia in critically ill comatose patients. A randomized study.
Authors:	Acquarolo A, Urli T, Perone G, Giannotti C, Candiani A, Latronico N.
Reference:	Intensive Care Med. 2005 Apr;31(4):510-6. Epub 2005 Mar 8.
Link:	Click here for a direct link to the paper. A password may be required for access to fulltext.
Abstract:	OBJECTIVE: To evaluate if a 3-day ampicillin-sulbactam prophylaxis can reduce the occurrence of early-onset pneumonia (EOP) in comatose mechanically-ventilated patients. DESIGN: This was a single-centre, prospective, randomised, open study. SETTING: A 10-bed general-neurological ICU in a 2,000-bed university hospital. PATIENTS AND PARTICIPANTS: Comatose mechanically-ventilated patients with traumatic, surgical or medical brain injury. INTERVENTIONS: Patients were randomized to either ampicillin-sulbactam prophylaxis (3 g every 6 h for 3 days) plus standard treatment or standard treatment alone. MEASUREMENTS AND RESULTS: Main outcome was the occurrence of EOP. Secondary outcome measures were occurrence of late-onset pneumonia, percentage of non-pulmonary infections and of emerging multiresistant bacteria, duration of mechanical ventilation and of ICU stay and ICU mortality. Interim analysis at 1 year demonstrated a statistically significant reduction of EOP in the ampicillin-sulbactam group, and the study was interrupted. Overall, 39.5% of the patients developed EOP, 57.9% in the standard treatment group and 21.0% in the ampicillin-sulbactam group (chi-square 5.3971; P =0.022). Relative risk reduction of EOP in patients receiving ampicillin-sulbactam prophylaxis was 64%; the number of patients to be treated to avoid one episode of EOP was three. No differences in other outcome parameters were found; however, the small sample size precluded a definite analysis. CONCLUSIONS: Antibiotic prophylaxis with ampicillin-sulbactam significantly reduced the occurrence of EOP in critically ill comatose mechanically ventilated patients. This result should encourage a large multicenter trial to demonstrate whether ampicillin-sulbactam prophylaxis reduces patient mortality, and whether antibiotic resistance is increased in patients receiving prophylaxis.
Are the Results Valid?
1. Was the assignment of patients to treatments randomized? ( Was allocation concealment maintained?)
	Yes, allocation concealment was maintained by the use of sealed opaque envelopes, which were kept in a locker, whose key was available to the attending intensivist.
2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
2a. Was followup complete?
	No (see fig 1). 150 patients were screened, 108 excluded (reasons given) and 42 patients were randomised. Four patients were randomised into the trial but 'excluded' from the intention-to-treat analysis due to 'protocol violations' (3 ventilated less than 24 hours, 1 had admission diagnosis of pneumonia. This is approximately 10% loss to follow-up.
2b. Were patients analyzed in the groups to which they were randomized?
	38 patients were analysed in the groups they were assigned to, the "drop outs" were not analysed in an intention to treat.
3. Were patients, health workers, and study personnel blind to treatment?
	NO. Patients in the standard treatment group did not receive a placebo. Treating physicians and paramedics were not blinded to treatment allocation; the outcome adjudicators (TU, CG) and the data analyst (NL) were.
4. Were the groups similar at the start of the trial?
	Yes, table 1 appears to be in balance.
5. Aside from the experimental intervention, were the groups treated equally?
	Not explictly stated. As there was not blinding to the treatment this cannot be guarenteed.
What are the Results?
1. How large was the treatment effect?
	This study was stopped early due to meeting the pre-specified Pocock stopping threshold at the 1 yr interim analysis (p = 0.024). The authors report that EOP was reduced from 57.9% to 21.0% (p = 0.022). The p-value reported above (0.022) is obtained from an uncorrected chi-square test. With continutity correction, which may be more appropriate in small samples, the p-value is 0.046. [calculated by Gord] The p-value obtained from Fisher's Exact test, which is definitely more appropriate when any cell in the 2x2 table is less than 5, is 0.045. [calculated by Gord using SAS] It is likely that the early stopping of this trial was inappropriate and subject to a Type I error. - Gord For an excellent review of stopping rules and type I errors, click here.
2. How precise was the estimate of the treatment effect?
	As calculated in the paper, the 95% CIs are: RRR=0.660 (0.127-0.875)95% CI ARR=0.368 (0.097-0.639) NNT = 2.7 (2-10)
Will the Results Help Me In Caring For My Patients?
1. Can the results be applied to my patient care?
	Of the 119 patients admitted to my hospital with similar Dx, 46 were ventilated >72 hours. 22 patients have positive trachial aspirates out of 37 available. Organism grown were; staph-9, H influ-10 Pneumococcus-3 others 6. Due to the small numbers in the trial, and dubious outcomes, more study is needed
2. Were all clinically important outcomes considered?
	Only EOP ( early onset pneumonia) was considered. Mortality outcomes needed plus others and financial secondary outcomes.
3. Are the likely treatment benefits worth the potential harms and costs?
	Connot tell.
What other people had to say about: Antibiotic prophylaxis of early onset pneumonia in critically ill comatose patients. A randomized study. Add Your Comment
[ Back ] Any questions or comments please contact Gordon S. Doig Implemented and designed by John Soer and Gordon Doig Page last modified on Friday August 10, 2001 www.EvidenceBased.net
	THIS IS A SPACE MESSAGE. THIS IS STILL THE BEST METHOD TO SPACE A TABLE FOR NETSCAPE AND INTERNET EXPLORER. IF YOU SEE THIS MESSAGE CONTACT WEBADMIN@EvidenceBased.net