H.K.F. van Saene

Significant Reduction of Faecal Endotoxin Pool by Oral Polymyxin E and Tobramycin in Human Volunteers

Healthy human volunteers carry a large amount of endotoxin in their intestines (a mean of 109 pg per gram of faeces) [1]. The controlling mechanisms preventing the occurrence of systemic endotoxaemia must be extremely effective, as even such small concentrations as 2–5 ng/kg of endotoxin in blood can cause fever, thrombocytopenia etc. The first barrier constituting this important control mechanism is an intact mucosal lining, through which leakage is minimal [2]. The second barrier is the liver; intestinal endotoxin which passes the intestinal mucosa (portal endotoxaemia) is effectively cleared by Kupffer cells in the liver [3]. Finally, the third line of defence against endotoxin is the endotoxin-binding properties of the blood which neutralize endotoxin following spillover of the liver [4].

Oral Vancomycin in the Control of MRSA Outbreaks in the ICU

Methicillin-resistant Staphylococcus aureus (MRSA) is a serious problem, both clinically and epidemiologically. Systemic vancomycin is the cornerstone for the therapy of invasive infections due to MRSA. However, vancomycin is potentially toxic and there is an increasing fear that microorganisms resistant to vancomycin such as vancomycin-resistant enterococci (VRE) [1] and vancomycin intermediate Staphylococcus aureus (VISA) will emerge [2].

Selective decontamination is not sticky.

Dear Editor-in-Chief, We read the case discussion by Dr. M. J. Smit et al. [1] reporting accumulation of oral antibiotics as an adverse effect of selective decontamination of the digestive tract. The article disappointed us as it did not describe in detail the composition of the paste or suspension nor did the authors estimate the incidence of their reported SDD side-effect. How can your readers assess what actually caused the observed problem if the constituents of the paste are not reported, in particular the percentage of the sticky compound carboxymethylcellulose (CMC) in the paste. The eradication of abnormal flora including aerobic Gram-negative bacilli (AGNB) and methicillinresistant Staphylococcus aureus (MRSA) from the oropharynx, i.e. oropharyngeal decontamination, is difficult. In 1981, Gerald Bodey wrote that ‘antibiotic prophylaxis was less effective against the flora of the throat, probably because of the short contact of antibiotic with the oral mucosa’ [2]. The solution came from the experience in dentistry where pastes and gels are commonly used for their prolonged contact time between salivary micro-organisms and metronidazole mixed with pastes and gels [3]. Stoutenbeek [4] was the first to decontaminate the oropharynx of ventilated patients using orabase paste mixed with 2% of polymyxin E, tobramycin and amphotericin B [4]. At the SDD meeting in Jersey in 1988, Crome recommended a gel containing 3% of CMC rather than a paste preparation for oropharyngeal decontamination, the main reason being that the paste is so adherent to the mucosa that removal is difficult and it has a drying effect on the mucosa [5]. In contrast, the gel has good adhesion to the mucosa with prolonged release of drugs but is easy to use. We have abandoned the paste and replaced it with a gel. The side effect described by Dr. Smit et al. has never been reported during the use of the 3% CMC gel over 20 years. The higher concentration of the CMC in the paste compared with the gel inhibits proper oral care allowing the patient to swallow residual buccally applied SDD paste sticking around the nasogastric tube in the oesophagus. Unfortunately, Smit et al. have not described the nasogastric tubes or intravenous cannulae they used nor their enteral feeding policies. We in Amsterdam change plastic devices in every patient every 7 days, and have never experienced ‘glued’ gastric tubes over the last 20 years during which time 14,000 patients received SDD. We believe that enteral feed is a significant cause of obstruction of the esophagus due to solidification of tube feeding [6]. As the total amount of paste used is too small to create the type of solid mass reported by Smit et al., we suggest that a more credible reason for the obstruction observed is that the amphotericin B suspension reacts with the gastric compounds and leads to the formation of a bezoar-type of mass in the upper gastro-intestinal tract following reflux.