Reply to comments on: Prophylactic antibiotics for burn patients
We thank Silvestri et al. (1) for bringing to our knowledge the
existence of another study that is eligible for inclusion in our meta-
analysis. We did not identify this study published in Annals of Burns and
Fire Disasters, which is not indexed in PubMed, nor in the other databases
included in our search or trial registry databases. A search of this
journal did not reveal further eligible trials. Two publications (2,3)
describe a study showing a highly significant reduction in all-cause
mortality with oral cotrimoxazole, non-absorbable colistin and nystatin
vs. no prophylaxis, among patients with total burn surface area (TBSA) of
25% or more in adults and 15% or more in children. The trial was quasi-
randomized, using patients’ admission number for randomization, non-
blinded and the number of randomized patients is unclear and different in
the two publications. We could not obtain further data from the authors,
thus the study is at high-risk for bias.
The term selective decontamination probably originates from early
studies attempting total decolonization of neutropenic cancer patients,
mainly in an attempt to prevent translocation of bacteria from the gut
(4). Currently, selective decontamination of the digestive tract (SDD)
refers to a short period of parenteral broad-spectrum antibiotic treatment
combined with longer enteral decontamination. Rather than referring to
this imprecise term (since there is nothing selective about this regimen)
and to acknowledge the fact that among burns patients parenteral
antibiotics also affect skin and integument, we separated interventions
into those including a systemic antibiotic, those limited to non-
absorbable antibiotics only and other local antibiotic prophylaxis (5). As
such, both de la Cal et al. and the new study enter the first category of
systemic prophylaxis. In the updated meta-analysis, (excluding the
allopurinol-prophylaxis study arm of the new study that does not fulfill
the review inclusion criteria), systemic antibiotic prophylaxis not
limited to the peri-operative setting, reduces all-cause mortality with a
risk ratio of 0.46 (95% confidence intervals 0.30 to 0.69), 6 trials, 469
patients, fixed effect model, with moderate heterogeneity (I2=40%).
Separating all systemic antibiotic prophylaxis trials into those with or
without a non-absorbable component shows that the advantage exists in both
subsets (Figure). Excluding Deutch 1998 that used only oral erythromycin
as the systemic antibiotic, the RR is 0.35 (95% CI 0.21 to 0.59) with no
Silvestri et al. stated that SDD is the only evidence-based maneuver
that significantly reduces mortality in critically ill patients, including
burns, without antimicrobial resistance emerging (6). We are more cautious
in our interpretation of the evidence among burn patients, given the high
risk of bias in some of the studies, such as the newly added trial. We
agree with Silvestri et al. that the intervention with the highest
potential for benefit is a combined enteral-parenteral regimen, such as
the one used in de la Cal et al. (7).
Bracco and Eggimann raised doubts as to the feasibility of conducting
a new randomized controlled trial (8). Although secondary outcomes are
important, antibiotic prophylaxis for burns patients will be accepted or
rejected only on the basis of its effect on mortality. All-cause mortality
encompasses the effects of resistance development on the individual
patient. In the 2007 National Burn Repository report (9), the average
mortality rate in burn injuries >20% TBSA (the inclusion criteria of
most included trials) between 1998 to 2007 was 23.5%. There were 526
children and 1,126 adults admitted with severe burns between 2005 to 2007
in Britain (10). The potential reduction of 50% all-cause mortality for
young people without an underlying disease does not leave much choice but
to encourage such a trial.
1. Luciano Silvestri, Hendrick K.F. van Saene, Pier Camillo Parodi,
and Miguel A. de la Cal. Selective decontamination of the digestive tract
reduces mortality in burn patients. Rapid response published in BMJ.com,
16 March 2010.
2. Abdel-Razek S.M., Abdel-Khalek A.H., Allam A.M., Shalaby H.,
Mandoor S., Higazi M. Impact of selective gastrointestinal decontamination
on mortality and morbidity in severely burned patients. Annals of Burns
and Fire Disasters 2000;13:213-215.
3. Shalaby HA, Higazi M, El Far N. Selective gastrointestinal
decontamination and burn wound sepsis. Annals of Burns and Fire Disasters
4. Schlesinger A, Paul M, Gafter-Gvili A, Rubinovitch B, Leibovici L.
Infection-control interventions for cancer patients after chemotherapy: a
systematic review and meta-analysis. Lancet Infect Dis 2009;9:97-107.
5. Avni T, Levcovich A, Ad-El DD, Leibovici L, Paul M. Prophylactic
antibiotics for burns patients: systematic review and meta-analysis. BMJ
2010;340:c241. doi: 10.1136/bmj.c24.
6. Silvestri L, de la Cal MA, Taylor N, van Saene HK, Parodi PC.
Selective decontamination of the digestive tract in burn patients: an
evidence-based maneuver that reduces mortality. J Burn Care Res 2010;
7. de La Cal MA, Cerda E, Garcia-Hierro P, van Saene HK, Gomez-Santos
D, Negro E, et al. Survival benefit in critically ill burned patients
receiving selective decontamination of the digestive tract: a randomized,
placebo-controlled, double-blind trial. Ann Surg 2005;241:424-30.
8. Bracco D, Eggimann P. Prophylaxis with systemic antibiotics in
patients with severe burns. BMJ 2010;340:c208. doi: 10.1136/bmj.c208.
9. Miller SF, Bessey P, Lentz CW, Jeng JC, Schurr M, Browning S; ABA
NBR Committee. National burn repository 2007 report: a synopsis of the
2007 call for data. J Burn Care Res 2008;29:862-70.
10. 2008 First iBID Report. UK National Burn Care Group. Available
Last accessed: 6 April, 2010.
All cause mortality in trials comparing systemic antibiotic
prophylaxis for burns patients, not limited to the peri-operative setting.
Studies are subcategorized by the presence of non-absorbable antibiotic in
the prophylaxis regimen and sorted by allocation concealment. (See
reference 5 for the studies included in the meta-analysis)
Competing interests: No competing interests