Selective digestive decontamination to prevent pneumonia after esophageal surgery.

We read with interest the review by Akutsu and Matsubara on the prevention of postoperative pneumonia in esophageal surgery. However, we believe that the section on oral care was not comprehensive because the use of selective decontamination of the digestive tract (SDD) was not accurately reported. Oropharyngeal flora has been recognized as the major source of potential pathogens causing pneumonia in critically ill patients requiring intensive care unit (ICU) treatment. SDD has been shown to prevent pneumonia developing during ICU treatment of nonoperated and operated patients. The concept of SDD is based on the observation that the eradication of the oropharyngeal and intestinal carriage of potential pathogens prevents serious infections, and mortality, in ICU patients. We were puzzled why Akutsu and Matsubara ignored the evidence of SDD as a preventive measure of pneumonia in esophagogastric surgery, citing only the negative study by Farran et al. In this Spanish randomized controlled study (RCT) evaluating the impact of SDD on anastomotic dehiscence and on pulmonary infection in esophagogastric surgery, the postoperative pneumonia was reduced, but not significantly. We performed a systematic review of all 60 published RCTs of SDD, and we found three RCTs assessing SDD in (gastro)-esophageal surgery (Table 1). A total of 410 patients (198 SDD, 212 control) were included. Fifty-six developed pneumonia, 15 (7.65%) in the SDD group and 41 (19.34%) in control. SDD significantly reduces the odds for pneumonia by 64% (odds ratio 0.36; 95% confidence interval 0.19-0.69; p = 0.0018). Interestingly, anastomotic leakage was significantly reduced in Shardey’s study to 2.9%, from 10.6% (p = 0.04), but in the Farran study, the reduction to 2.5% from 5.9% was not significant. Remarkably, there is a substantial difference in the decontaminating agents used in the three RCTs. The Spanish study employed erythromycin rather than polymyxins, which were used in the other two studies. 5) In comparison with polymyxins, macrolides, such as erythromycin, are inferior decontaminating agents because they are absorbed from the gastrointestinal tract, and their spectrum of activity covers only “normal” potential pathogens. Erythromycin has been shown to clear oropharyngeal carriage of Streptococcus pneumoniae and Haemophilus influenzae and gut carriage of Escherichia coli following salivary and biliary excretions, but it failed to eradicate “abnormal” potential pathogens such as aerobic gram-negative bacilli. We believe these observations may help the reader to recognize that SDD may be considered as a preventive maneuver that significantly reduces pneumonia after esophageal surgery.