Hendrik K.F. van Saene

Impact of selective decontamination of the digestive tract on fungal carriage and infection: systematic review of randomized controlled trials

ObjectiveTo determine the impact of the antifungal component of selective decontamination of the digestive tract on fungal carriage, infection and fungaemia.DesignMeta-analysis of randomized controlled trials of selective decontamination of the digestive tractStudy selectionData sources included Medline, Embase, Cochrane Register of Controlled Trials, previous meta-analyses, personal communications and conference proceedings, without restriction of language or publication status. All randomized trials were selected that compared oropharyngeal and/or intestinal administration of antifungals amphotericin B or nystatin, as part of selective decontamination protocol, with no treatment in the controls. There were 42 randomized controlled trials with a total of 6,075 critically ill patients.MethodsThree reviewers independently applied selection criteria, performed quality assessment and extracted the data. The main outcome measures were patients with fungal carriage, patients with fungal infections and patients with fungaemia. Odds ratios were pooled with the random effect model.Measurements and resultsEnteral antifungals significantly reduced fungal carriage (odds ratio 0.32, 95% confidence interval 0.19–0.53) and overall fungal infections (0.30, 0.17–0.53). Fungaemia was not significantly reduced in the treatment group (0.89, 0.16–4.95).ConclusionsAntifungals, as part of selective decontamination of the digestive tract, reduce fungal carriage and infection but not fungaemia in critically ill patients and may justify the inclusion of an antifungal component in the decontamination protocol.

Selective intestinal decontamination in advanced chronic heart failure: a pilot trial.

Endotoxin, derived from intestinal aerobic Gram‐negative bacilli (AGNB), could be an important monocyte activator in chronic heart failure (CHF). The effect of selective decontamination of the digestive tract (SDD) on intracellular monocyte cytokine production, monocyte CD14 expression, circulating endotoxin and cytokines, and flow‐mediated dilation (FMD) was studied in patients with severe CHF.

Effect on Colonization Resistance: an Important Criterion in Selecting Antibiotics

Infections in humans are most often caused by aerobic microorganisms colonizing the digestive tract. Aerobic microorganisms are constantly entering the digestive tract with food, but colonization is resisted by autochthonous anaerobic flora (microbial colonization resistance) and by host-related factors (physiologic colonization resistance). Antibiotics to which the autochthonous anaerobic flora are sensitive and that achieve sufficiently high concentrations at the sites of colonization will reduce colonization resistance. Consequently, resistant aerobic flora of the digestive tract may reach high concentrations, increasing the risk of superinfection. Therefore, when choosing antimicrobial agents for therapy, the effect on colonization resistance should be taken into account. Immunosuppressed hosts and acutely ill patients undergoing mechanical ventilation can be protected from serious infections by eliminating the most dangerous species of the aerobic endogenous flora, leaving colonization resistance intact. This is called selective decolonization. This article summarizes the effects of antimicrobial agents on colonization resistance.

Is Selective Digestive Decontamination Useful in Controlling Aerobic Gram-Negative Bacilli Producing Extended Spectrum Beta-Lactamases?

AIMS To identify outbreak episodes of either carriage or infection due to extended spectrum beta-lactamases producing aerobic Gram-negative bacilli (AGNB-ESBL); to establish whether AGNB-ESBL, sensitive to tobramycin, become resistant over time; and to evaluate the impact of selective decontamination of the digestive tract (SDD) on abnormal carriage of AGNB-ESBL. DESIGN AND SETTING All children admitted to the pediatric intensive care unit (PICU) over a 12-month period had biweekly surveillance cultures of throat and rectum and diagnostic cultures when clinically indicated. All AGNB were tested for ESBL, and the positive isolates were sent for molecular typing. The PICU uses SDD (parenteral cefotaxime and enteral polymyxin E/tobramycin) to control abnormal carriage. Patients who had at least one AGNB-ESBL were included in the study. RESULTS During the study period, 1,101 children were admitted to the PICU. There were 39 patients (3.5%) with a total of 236 cultures positive for AGNB-ESBL. Twenty-eight patients (2.5%) were carriers, and 11 (1%) had proven infections. Organisms isolated from the first culture were 14 patients with Klebsiella pneumoniae, 8 with Enterobacter cloacae, 7 with Citrobacter freundii, 5 with Klebsiella oxytoca, and 5 with Escherichia coli. In the first sample, 59% of isolates showed tobramycin resistance. Molecular typing confirmed that there were five different strains of K. pneumoniae and that similar strains were not isolated in the same period. CONCLUSIONS SDD is an effective measure to control AGNB-ESBL and to avoid outbreak episodes of either carriage or infection. When tobramycin resistance is found, replacing it with another aminoglycoside based on antibiogram may be more effective in achieving AGNB clearance.

Selective decontamination of the digestive tract: rationale behind evidence-based use in liver transplantation.

Selective decontamination of the digestive tract (SDD) is a prophylactic strategy designed to prevent or minimize the impact of endogenous infections by potentially pathogenic microorganisms (PPM) in patients at high risk of infection.1 The purpose of SDD is to prevent or eradicate, if initially present, oropharyngeal and gastrointestinal carriage of PPM, especially aerobic gram-negative bacilli (AGNB), but also Staphylococcus aureus and yeasts, leaving the indigenous flora, which are thought to play a role in the resistance to colonization by PPM, predominantly undisturbed.2 The overall aim is a reduction in morbidity and mortality. The physiological phenomenon that the normal, mainly anaerobic flora is required to control the abnormal aerobic PPM was recognized early, after the introduction of antimicrobial agents.3 Antibiotics active against anaerobes and excreted in the gut may suppress the normal indigenous flora. They disregard the ecology and may cause candidiasis of mouth, vagina, and groin.4 These flora-suppressing antimicrobials promote yeast overgrowth, defined as 10 colony-forming units per milliliter of saliva and per gram of feces, following the excretion of microbiologically active antibiotic concentrations into the throat and/or gut via saliva, bile, and mucus. The need to preserve the normal indigenous flora was also acknowledged for controlling overgrowth of S. aureus5 and AGNB.6 Previous antibiotic administration lowers the infecting doses of high-level enteric pathogens Salmonella7 species and Clostridium difficile.8 More recently, overgrowth of low-level pathogens including vancomycin-resistant enterococci (VRE) is prevented by antimicrobials that respect the gut ecology.9 In 1971, van der Waaij quantified the physiological phenomenon of the normal flora controlling the abnormal flora using challenge experiments in mice.2 He defined colonization resistance as the concentration of the bacterial challenge strain expressed by the log of colony-forming units per milliliter required to result in abnormal carriage in half the animals. Generally, healthy animals possess a high colonisation resistance of 9 as they clear high doses of 10 AGNB, including Pseudomonas aeruginosa, Klebsiella pneumoniae and Enterobacter cloacae, contaminating their drinking water. Antimicrobials including cephradine and cefotaxime did not promote the establishment of abnormal flora and were labelled as ecology-friendly, or “green,” antibiotics.10 The abnormal carrier state was established in 50% of the animals receiving antibiotics like ampicillin and flucloxacillin after challenging them with 10 PPM. These agents decreased the resistance of mice to colonization to 5 and were considered to be “red,” as they disregard the animal gut ecology. Amoxycillin was found to be “amber,” as only high doses lowered the colonization resistance of mice. These antimicrobials were subsequently tested in healthy volunteers also in challenge studies.11–13. Vollaard demonstrated that none of the antimicrobials were found to be completely ecology friendly.13 They invariably impacted colonization resistance. He argued that the gut ecology is an extremely fragile balance and very susceptible to antimicrobial agents. However, there were still major differences among antimicrobial agents in terms of their influence on the indigenous flora. In the volunteer studies, the disregard of ampicillin and amoxycillin for the ecology was significantly worse compared with that of cephradine and cefotaxime. Abnormal carriage with ampicillin and amoxycillin was more frequent and lasted longer compared with cephraAbbreviations: SDD, selective decontamination of the digestive tract; PPM, potentially pathogenic microorganisms; AGNB, aerobic gram-negative bacilli; VRE, vancomycin-resistant enterococci; ICU, intensive care unit; RCTs, randomized controlled trials; MRSA, methicillin-resistant Staphylococcus aureus. From the Department of Medical Microbiology, University of Liverpool, UK, and Department of Intensive Care, Onze Lieve Vrouwe Gasthius Hospital, Amsterdam, The Netherlands. Address reprint requests to HKF van Saene, M.D., Department of Medical Microbiology, University of Liverpool, Duncan Building, Daulby Street, Liverpool L69 3GA, United Kingdom. Telephone: 44-151-2525006; FAX: 44-151-2525356; E-mail: Rick.VanSaene@ RLCH-TR.NWEST.NHS.UK Copyright

Pulmonary bacterial co-infection in children ventilated for severe respiratory syncytial virus bronchiolitis is common

Sir, In a retrospective study on concurrent bacterial infection in infants ventilated for respiratory syncytial virus (RSV), bronchiolitis Kneyber et al. concluded that the “improvement in the diagnosis of a pulmonary concurrent bacterial infection is warranted to reduce the overuse of antimicrobial drugs” in this patient group [1]. This issue was highlighted by Andrews et al. in “Intensive Care Medicine’s Year in Review” 2005 article [2] as “Only a minority had evidence of bacterial infection”. Our view differs and we question this opinion/interpretation. A prospective study of 165 consecutive children in whom an admission broncho-alveolar lavage (BAL) sample was obtained found that up to 40% of patients admitted with severe RSV bronchiolitis were infected with bacteria in their lower airways [3]. This is in keeping with Kneyber et al. who found that 33% of children (9 of 24) in whom admission endotracheal aspirates were performed had a positive bacterial culture [1]. Duttweiler et al. retrospectively studied 127 infants admitted to intensive care for RSV bronchiolitis and found that 44% of those ventilated and endotracheally sampled had “concomitant bacteria pneumonia” [4]. Likewise, Randolph et al. retrospectively examined 165 previously healthy infants admitted to the intensive care unit with laboratoryconfirmed RSV infection and found that up to 38% of the intubated infants had “probable” or “possible” bacterial pneumonia [5]. Pulmonary bacterial co-infection is common in children ventilated for severe RSV bronchiolitis. The statement, “Only a minority had evidence of bacterial infection” [2] underscores the incidence in this patient group. Kneyber et al. [1] and Thorburn et al. [3] acknowledged that their studies may even have underestimated the actual number of concurrent bacterial infections. All the patients in both these studies [1, 3] received concomitant antibiotics, which limits any interpretation on the influence of antibiotics on outcome. This also limits comment on the “overuse of antimicrobials” [1, 2]. The only way to prove a definite impact of antimicrobials in this study population would be to carry out a randomized control trial. In the interim, one could argue that a bacterial co-infection rate of up to 40% in ventilated RSV patients is sufficient to warrant prophylactic/empirical antibiotics on admission to the paediatric intensive care unit (PICU). These critically ill children run a serious risk of developing bacterial pneumonia [6]. While awaiting further studies, we think that empirical antibiotic therapy appears to be justified in children with severe RSV bronchiolitis, pending the results of definitive cultures. “Improvement in the diagnosis of a pulmonary concurrent bacterial infection” can be achieved by obtaining lower airways secretions on admission to PICU in all patients. In the absence of finding bacteria in the lower airways, antibiotics can subsequently be stopped. References

Digestive decontamination in burn patients: A systematic review of randomized clinical trials and observational studies

OBJECTIVE The objective of this systematic review is to assess the effect of selective digestive decontamination (SDD) or non-absorbable enteral antibiotics (EA) on mortality, the incidence of infection and its adverse effects in burn patients. MATERIAL AND METHODS Systematic review of randomized clinical trials (RCT) or observational studies enrolling burn patients, and comparing SDD or EA prophylaxis with placebo or no treatment. The search includes Pubmed/Medline, EMBASE, WOS, Cochrane Library (1970-2015). Bibliographic references were also reviewed, as well as communications presented at conferences (2012-2015), without language restrictions. Two reviewers inspected each reference identified by the search independently; the risk of bias was assessed with the Cochrane Collaboration method for RCT and the Newcastle Ottawa Scale for observational studies. RESULTS Five RCT and 5 observational studies were identified enrolling a total of 1680 patients. The overall methodological quality of the studies was poor. The pooled effect of RCT using EA was OR: 0.62 (95% CI: 0.20-1.94). The only RCT using SDD reported OR 0.20 (95% CI: 0.09-0.81). The incidence of Enterobacteriaceae bloodstream was lower in cases treated with SDD or EA. The incidence of pneumonia was only reduced in the studies using SDD. None of the studies reported an increase in antibiotic resistance but in one RCT SDD was associated to an increase in methicillin-resistant Staphylococcus aureus infections, that was controlled with enteral vancomycin. CONCLUSIONS SDD and EA have shown a beneficial effect in burn patients. Both practices are safe. Higher quality RCTs should be conducted to properly assess the efficacy and safety of SDD in this population.

Infections and Antibiotic Therapy in Surgical Newborn Infants

Colonisation of the gastro-intestinal tract of newborn infants starts immediately after birth and occurs within a few days. Initially, the type of delivery (passage through the birth canal versus caesarean section) and the type of diet (breast versus formula feeding) might affect the colonisation pattern. Nearly all full-term, formula-fed, vaginally delivered infants were colonised with anaerobic bacteria within 4–6 days. 61% harboured Bacteroides fragilis. In contrast, anaerobes were present in 59% and B. fragilis in only 9% of infants delivered by caesarean section, suggesting that significant contamination occurred during passage through the birth canal. Both prematurity and breast feeding reduced the likelihood of isolating anaerobic species. Enterococci were isolated from all neonates, Escherichia coli from 82.6%, anaerobic cocci from 52.2% and both streptococci and staphylococci from 34.8%. Colonisation of the small bowel occurs perorally. In newborn infants with congenital small bowel obstruction, a faecal-type flora is found immediately proximal to the site of obstruction, and the distal bowel remains sterile.

The addition of enteral to parenteral antimicrobials may prolong antibiotic era

Resistance to parenteral antimicrobials generally occurs within two years after introduction into general use. The site where de novo resistance develops has been acknowledged to be the gut. Overgrowth of abnormal flora, defined as 105 potential pathogens per g of faeces is a risk factor for resistance following increased spontaneous mutation leading to polyclonality and antimicrobial resistance. As parenteral antimicrobials generally fail to eradicate the abnormal carrier state in overgrowth concentrations due to sub-lethal concentrations in bile and mucus the enteral antimicrobials polymyxin/tobramycin aiming at converting the abnormal carrier state into normal carriage, are the essential component of selective decontamination of the digestive tract (SDD), because they eradicate carriage and overgrowth including resistant mutants, maintaining the usefulness of parenteral antimicrobials.