An Antibiotic Past May Save Lives at the ICU.

16 03 2009


Respiratory tract infections acquired in the intensive care unit (ICU) are important causes of morbidity and mortality, the most significant risk factor being mechanical ventilation. It is thought that hospital pneumonia commonly originates from flora colonized in the patient’s oropharynx (the area of the throat at the back of the mouth). Therefore, reduction of respiratory tract infections has been obtained by putting patients in semirecumbent instead of supine position. Another approach is selective decontamination. There are two methods of selective decontamination, SDD and SOD.

  1. SDD, Selective Decontamination of the Digestive tract consists of the administration of topical nonabsorbable antibiotics in the oropharynx and gastrointestinal tract, often concomitant with systemic antibiotics. It aims to reduce the incidence of pneumonia in critically ill patients by diminishing colonization of the upper respiratory tract with aerobic gram-negative bacilli and yeasts, without disrupting the anaerobic flora.
  2. SOD, Selective Oropharyngeal Decontamination is application of local antibiotics in the oopharynx only.

Both approaches were first introduced in the Netherlands. Most trials suggested that SDD lowered infection rates, but lacked statistical power to demonstrate an effect on mortality. However, meta-analyses and three single-center, randomized studies, did show a survival benefit of SDD in critically ill patients. Several studies had suggested that the local variant, SOD, was also effective, but SOD was never directly compared with SDD in the same study. Because of methodological issues and concern about increasing antibiotic resistance the use of both SDD and SOD has remained controversial. Even in the Netherlands where guidelines recommended the use of SDD after a Dutch publication in the Lancet (de Jonge et al, 2003) had shown the mortality to drop with 30% in the Academic Medical Center in Amsterdam, only 25% of the emergency doctors followed the guidelines.

The present Dutch study, published in the NEJM (2009), was undertaken to determine the effects on mortality in a head to head comparison of SDD and SOD. The effectiveness of SDD and SOD was determined in a crossover study using cluster randomization in 13 Dutch ICU’s, differing in size and teaching status. Cluster randomization means that ICU’s rather than the individual patients were randomized to avoid that one treatment regimen would influence the outcome of another regimen. Crossover implies that all three treatments (SDD, SOD, standard care) were administered in a random order in all ICU’s.

A total of 5939 patients were enrolled in this large study. Patients were eligible if they were expected to be intubated for more than 48 hours or to stay in the ICU for more than 72 hours. The SDD regimen involved four days of intravenous cefotaxime along with topical application of tobramycin, colistin and amphotericin B; the SOD regimen used only the topical antibiotics. Both regimens were compared with standard care. The duration of the study was six months, and the primary end point was 28-day mortality.

Of the 5,939 patients, 1,990 received standard care, 1,904 received SOD and 2,405 received SDD. Crude mortality rates in the three groups were 27.5%, 26.6% and 26.9%, respectively. These differences are not very huge and benefit was only discernable after adjustment for covariates (age, sex, APACHE II score, intubation status, medical specialty, study site, and study period): adjusted* odds ratios for 28-day mortality were 0.86 (95% CI, 0.74 to 0.99) in the SOD group and 0.83 (95% CI, 0.72 to 0.97) in the SDD group compared with standard care. This corresponded with the needed-to-treat numbers (NNT’s) of 29 and 34 to prevent one casualty at day 28 for SDD and SOD, respectively.

The limitations of the study (acknowledged by the authors) were the absence of concealment of allocation (due to the study design it was impossible to conceal the allocation for doctors at the wards), differences at baseline between the standard care and treatment groups and a mismatch between the original analysis plan and the study design (originally specified in-hospital death was the primary end point, but this did not take into account analysis of cluster effects.)

Selective Decontamination also improved microbiological outcomes, such as carriage of gram-negative bacteria in the respiratory and intestinal tracts and ICU-acquired bacteriemia. During the study periods the prevalence rates for antibiotic-resistant gram-negative bacteria were lower in the SOD and SDD periods than during the standard-care periods.

The authors concluded that both SDD and SOD were effective compared with standard care. Given the similarity in effects on survival between the treatment groups, the SOD regimen seems preferable to the SDD regimen, becauses it minimizes the risk of antibiotic resistance which poses a major threat to patients admitted to ICU’s. It should be noted that MRSA-infections are very rare in the Netherlands and in Scandinavia. The outcome of the study might therefore be different after long term treatment and/or in regions with a high prevalence of MRSA.


ResearchBlogging.orgde Smet, A., Kluytmans, J., Cooper, B., Mascini, E., Benus, R., van der Werf, T., van der Hoeven, J., Pickkers, P., Bogaers-Hofman, D., van der Meer, N., Bernards, A., Kuijper, E., Joore, J., Leverstein-van Hall, M., Bindels, A., Jansz, A., Wesselink, R., de Jongh, B., Dennesen, P., van Asselt, G., te Velde, L., Frenay, I., Kaasjager, K., Bosch, F., van Iterson, M., Thijsen, S., Kluge, G., Pauw, W., de Vries, J., Kaan, J., Arends, J., Aarts, L., Sturm, P., Harinck, H., Voss, A., Uijtendaal, E., Blok, H., Thieme Groen, E., Pouw, M., Kalkman, C., & Bonten, M. (2009). Decontamination of the Digestive Tract and Oropharynx in ICU Patients New England Journal of Medicine, 360 (1), 20-31 DOI: 10.1056/NEJMoa0800394

de Jonge E, Schultz M, Spanjaard L, et al. Effects of selective decontamination of the digestive tract on mortality and acquisition of resistant bacteria in intensive care: a randomised controlled trial. Lancet 2003;362:1011-1016 (PubMed citation)

Wim Köhler (2009) Smeren tegen infectie, NRC Handelsblad, Wetenschapsbijlage 3,4 januari (Dutch, online)

Barclay, L & Vega, C (2009) Selective Digestive, Oropharyngeal Decontamination May Reduce Intensive Care Mortality, Medscape

File, T.M., Bartlett J.G.,& Thorner, A.R. Risk factors and prevention of hospital-acquired (nosocomial); ventilator-associated; and healthcare-associated pneumonia in adults.www.uptodate)

Photo Credit (CC): (JomCleay)