Eradication of endotracheal tube biofilm by nebulised gentamicin.

OBJECTIVE: To compare the efficacy of gentamicin, nebulised via the endotracheal tube (ET), with that of parenteral cefotaxime or parenteral cefuroxime in preventing the formation of ET biofilm. SETTING: General intensive care units in two university teaching hospitals. DESIGN: The microbiology of ET biofilm from 36 ICU patients eligible to receive antibiotic prophylaxis was examined. Peak and trough tracheal concentrations of gentamicin, cefotaxime or cefuroxime were measured in each patient group, on the 2nd day of intubation. PATIENTS: Twelve patients received gentamicin (80 mg) nebulised in 4 ml normal saline every 8 h, 12 cefotaxime (1 g, 12 hourly) and 12 cefuroxime (750 mg, 8 hourly). Prophylaxis was continued for the duration of intubation. MEASUREMENTS AND RESULTS: Samples of tracheal secretions were taken on the 2nd day of ventilation for determination of antibiotic concentrations. Following extubation, ETs were examined for the presence of biofilm. Pathogens considered to be common aetiological agents for VAP included Staphylococcus aureus, enterococci, Enterobacteriaceae and pseudomonads. While microbial biofilm was found on all ETs from the cephalosporin group, microbial biofilm of these micro-organisms was found on 7 of the 12 ET tubes from patients receiving cefotaxime [ S. aureus (4), pseudomonads (1), Enterobacteriaceae (1), enterococcus (1)] and 8 of the 12 ET tubes from patients receiving cefuroxime [Enterobacteriaceae (6), P. aeruginosa (1) and enterococcus (1)]. While microbial biofilm was observed on five ETs from patients receiving nebulised gentamicin, none of these were from pathogens for ventilator-associated pneumonia (VAP). Tracheal concentrations of both cephalosporins were lower than those needed to inhibit the growth of pathogens recovered from ET tube biofilm. The median (and range) concentrations for cefotaxime were 0.90 (<0.23-1.31) mg/l and 0.28 (<0.23-0.58) mg/l for 2 h post-dose and trough samples, respectively. Two hours post-dose concentrations of cefuroxime (median and range) were 0.40 (0.34-0.83) mg/l, with trough concentrations of 0.35 (<0.22-0.47) mg/l. Tracheal concentrations (median and range) of gentamicin measured 1 h post-nebulisation were 790 (352-->1250) mg/l and then, before the next dose, were 436 (250-1000) mg/l. CONCLUSION: Nebulised gentamicin attained high concentrations in the ET lumen and was more effective in preventing the formation of biofilm than either parenterally administered cephalosporin and therefore may be effective in preventing this complication of mechanical ventilation.

The concomitant development of poly(vinyl chloride)-related biofilm and antimicrobial resistance in relation to ventilator-associated pneumonia.

Ventilator-associated pneumonia is a major cause of death in intensive care patients and the endotracheal tube, commonly fabricated from poly(vinyl chloride) (PVC), is acknowledged as a significant factor in this. Bacteria colonise the biomaterial, thereby adopting a sessile mode of growth that progresses to the establishment of an antibiotic-resistant biofilm by the accretion of a protective glycocalyx. This study examined the sequential steps involved in the formation of biofilm on PVC by atomic force microscopy and the concomitant development of resistance to an antibiotic (ceftazidime) and to a non-antibiotic antimicrobial agent (hexetidine). Staphylococcus aureus and Pseudomonas aeruginosa isolated from ET tube biofilm were employed. The surface microrugosity of bacteria growing in sessile mode on PVC decreased significantly (p < 0.05) over the period 4, 24, 48 h and 5 d. The progressive accretion of bacterial glycocalyx was readily visualised in micrographs leading to a smoother surface topography with time. The minimum inhibitory concentrations (MIC) and minimum bactericidal concentrations (MBC) for ceftazidime and hexetidine against planktonic (suspension) S. aureus were lower than for Ps. aeruginosa. Furthermore, sessile populations of S. aureus and Ps. aeruginosa on PVC exhibited greater resistance to both ceftazidime and hexetidine when compared to planktonic bacterial growth. The efficacy of the agents, determined by kill kinetics, against sessile bacteria was dependent on age, with established biofilms (> or = 24 h) significantly more resistant (p < 0.05) than early sessile populations (< or = 4 h). Importantly, for practice, even newly colonised bacteria (1 h) were significantly more resistant to antibiotic than planktonic bacteria. Hexetidine was significantly more active (p < 0.05) than ceftazidime on biofilms of both isolates, irrespective of age, with total kill within 24 h treatment. Hexetidine may offer promise in the resolution of infection associated with PVC endotracheal tubes.

Conditioning film and environmental effects on the adherence of Candida spp. to silicone and poly(vinylchloride) biomaterials.

The reported incidence of colonization of oropharyngeal medical devices with Candida spp. has increased in recent years, although few studies that have systematically examined the adherence of yeast cells to such biomaterials, the primary step in the process of colonization. This study, therefore, examined the effects of oropharyngeal atmospheric conditions (5% v/v carbon dioxide) and the presence of a salivary conditioning film on both the surface properties and adherence of Candida albicans, Candida krusei and Candida tropicalis to PVC and silicone. Furthermore, the effects of the salivary conditioning film on the surface properties of these biomaterials are reported. Growth of the three Candida spp. in an atmosphere containing 5% v/v CO2 significantly increased their cell surface hydrophobicity and reduced the zeta potential of C. albicans and C. krusei yet increased the zeta potential of C. tropicalis (p<0.05). Furthermore, growth in 5% v/v CO2 decreased the adherence of C. tropicalis and C. albicans to both PVC and silicone, however, increased adherence of C. krusei (p<0.05). Pre-treatment of the microorganisms with pooled human saliva significantly decreased their cell surface hydrophobicity and increased their adherence to either biomaterial in comparison to yeast cells that had been pre-treated with PBS (p<0.05). Saliva treatment of the microorganisms had no consistent effect on microbial zeta potential. Interestingly, adherence of the three, saliva-treated Candida spp. to saliva-treated silicone and PVC was significantly lower than whenever the microorganisms and biomaterials had been treated with PBS (p<0.05). Treatment of silicone and PVC with saliva significantly altered the surface properties, notably reducing both the advancing and receding contact angles and, additionally, the microrugosity. These effects may contribute to the decreased adherence of saliva-treated microorganisms to these biomaterials. In conclusion, this study has demonstrated the effects of physiological conditions within the oral cavity on the adherence of selected Candida spp. to biomaterials employed as oropharyngeal medical devices. In particular, this study has ominously shown that these materials act as substrates for yeast colonization, highlighting the need for advancements in biomaterial design. Furthermore, it is important that physiological conditions should be employed whenever biocompatibility of oropharyngeal biomaterials is under investigation.

Implications of endotracheal tube biofilm for ventilator-associated pneumonia.

OBJECTIVE: To determine the relationship between, and antibiotic resistance of, endotracheal tube (ET) biofilm and pulmonary pathogens in ventilator-associated pneumonia (VAP). SETTING: General intensive care units in two university teaching hospitals. DESIGN: The microbiology of ET biofilm and tracheal samples from patients with and without VAP were compared. For individual patients, matching pairs of pathogens were confirmed as identical and characterised for antibiotic susceptibility. PATIENTS: 40 intensive care unit patients - 20 with VAP, 20 without VAP as control. The duration of intubation (median and range) was 6.5 days (3-17) and 5 days (2-10), respectively. MEASUREMENTS AND RESULTS: Samples of tracheal secretions were taken during ventilation for bacteriological culture. Following extubation, ETs were examined for the presence of biofilm. Isolates of high pathogenic potential included Staphylococcus aureus, enterococci, Enterobacteriaceae, pseudomonads and Candida spp. Where the same microorganism was found on tracheal and ET samples by phenotyping, these were confirmed as identical by genotyping and characterised for antibiotic susceptibility in both the free floating and biofilm forms. Seventy per cent of patients with VAP had identical pathogens isolated from both ET biofilm and tracheal secretions. No pairing of pathogens was observed in control patients (p < 0.005). Susceptibility data for these pairs show that the ET acts as a reservoir for infecting microorganisms which exhibit significantly greater antibiotic resistance than their tracheal counterparts. CONCLUSION: This investigation provides further evidence for the role of ET biofilm in VAP. The difficulty in eradicating an established microbial biofilm using antibiotics implies that increased attention must be directed towards modification of the ET to prevent or substantially reduce biofilm formation.

Conditioning fluid influences on the surface properties of silicone and polyurethane peritoneal catheters: implications for infection.

Catheter-related infection remains a considerable problem in continuous ambulatory peritoneal dialysis (CAPD). This study examined the adherence of clinical isolates of Staphylococcus epidermidis to commercially available polyurethane and silicone peritoneal catheters in the presence and absence of a proteinaceous conditioning film. In addition, the effects of the conditioning film on the surface properties (advancing and receding contact angles, and surface rugosity) of these biomaterials were investigated. Bacterial adherence to polyurethane and silicone catheters, pre-treated with phosphate-buffered saline (PBS) or artificial spent peritoneal dialysate (ASD) for 1 h at 37 degrees C, was examined using a radiometric (2-3H-adenine) adherence assay. The advancing and receding contact angles and the surface rugosity of ASD- and PBS-treated biomaterials were examined using a dynamic contact angle analyser and an atomic force microscope, respectively. The bacterial isolates were selected to represent high and low cell surface hydrophobicity. The hydrophobic isolate exhibited both a significantly greater rate and a significantly greater extent of adherence than the hydrophilic isolate to both catheter materials, independent of pre-treatment. In general, pre-treatment of the catheter materials with ASD significantly decreased the subsequent adherence of both isolates owing to the deposition of a conditioning film on the surface of the biomaterial. ASD treatment also decreased both the advancing and receding contact angles and the surface rugosity of both catheter materials. This study highlights the influence of both bacterial cell surface hydrophobicity and biomaterial surface conditioning films on bacterial adherence to CAPD catheters. In addition, it is recommended that the effects of proteinaceous conditioning films on biomaterial surface properties should be considered when assessing materials for medical devices and products.

Incidence and nature of peritoneal catheter biofilm determined by electron and confocal laser scanning microscopy.

Thirty-two Tenckhoff catheters retrieved from continuous ambulatory peritoneal dialysis patients with a history of peritonitis were examined for microbial biofilm. Confocal laser scanning microscopy was successfully employed to visualize bacteria in biofilm occluded from view by scanning electron microscopy. Occluded but viable microbial biofilm was associated with 17 (81%) catheters from patients free from infection following renal transplant. Mixed isolate biofilm with two or more isolates of coagulase-negative staphylococci or Staphylococcus aureus was found on 41% of these catheters. Clearly visible viable biofilm consisting exclusively of Pseudomonas aeruginosa occurred on all four catheters removed due to recurrent peritonitis. Five (71%) catheters retrieved from patients transferred to haemodialysis had viable biofilm. Antibiotic sensitivities of the biofilm isolates were similar in profile to those reported for non-biofilm isolates from infected dialysate. Persistence of catheter biofilm despite direct contact with therapeutic levels of antibiotics in peritoneal dialysate requires that attention be directed towards improving antibiotic efficacy against peritonitis-causing bacteria in biofilm form.

Interaction of sucralfate with antibiotics used for selective decontamination of the gastrointestinal tract.

The interactions of sucralfate with colistin sulfate, with tobramycin sulfate, and with amphotericin B were studied. Sucralfate 500 mg was added to 40 mL of distilled water adjusted to pH 3.5 with hydrochloric acid. Stock solution of one of the three antibiotics was added to give a final colistin concentration of 50 mg/L (as the sulfate salt), final tobramycin concentration of 50 mg/L (as the sulfate salt), and final amphotericin B concentration of 25 mg/L. Samples were removed from each sucralfate-antibiotic mixture at 0, 5, 10, 15, 30, 45, 60, and 90 minutes and analyzed for antibiotic concentration by high-performance liquid chromatography (colistin), enzyme immunoassay (tobramycin), and spectrophotometry (amphotericin B). To determine if any interaction was reversible, the mixtures were stored for 90 minutes without sampling, the pH was adjusted to 6.5-7.0, and samples were removed and analyzed. All tests were performed in triplicate, and the temperature was maintained at 25 degrees C. Significant drug loss was observed starting at five minutes for each antibiotic-sucralfate mixture. This effect was not reversible in the less acidic environment. The concentrations of colistin, tobramycin, and amphotericin B declined rapidly when each drug was combined separately with sucralfate.

Confocal laser scanning microscopy of peritoneal catheter surfaces.

Surface topography of used (in situ > 12 months) and unused CAPD catheters was studied by scanning electronmicroscopy (SEM) and confocal laser scanning microscopy (CLSM). Microbial biofilm was observed on all used catheters. Disruption and removal of the attached biofilm revealed extensive pitting of the catheter surface and scoring within the catheter pores. Similar, though less extensive, surface defects were present in unused catheters. Examination by CLSM, with software specific to the determination of surface topography, showed used catheters to have a surface microrugosity greater than that of unused catheters (p < 0.0005). Adherence studies with radiolabelled Staphylococcus epidermidis demonstrated increased adherence to used than to unused catheters (p < 0.0005) after 48 h. However, when catheters were pre-treated with spent dialysate there was a substantial reduction in bacterial adherence to either catheter and no significant difference in adherence to used and unused catheters. Surface microrugosity of CAPD catheters increases during use but is unlikely to be an important factor in bacterial adherence in vivo.

Selective decontamination of the digestive tract (SDD) does not prevent the formation of microbial biofilms on endotracheal tubes.

The purpose of selective decontamination of the digestive tract (SDD) is to eradicate potentially disease-producing micro-organisms from the oropharynx and gastro-intestinal tract of intensive care unit (ICU) patients, thereby reducing the incidence of nosocomial sepsis, particularly pneumonia. Microbial biofilms form on endotracheal (ET) tubes even when SDD is being administered and may represent a persistent focus for infection. The aim of this investigation was to determine the susceptibilities of organisms adherent to ET tubes to SDD antibiotics (amphotericin B, tobramycin and polymyxin) and to measure the concentrations of these agents in the tracheal aspirates of 11 patients who were being mechanically ventilated. Following extubation, a section was cut from the tip of each ET tube and any adherent microorganisms subsequently isolated were identified and their MICs determined. Samples of tracheal aspirate were obtained three hours after administration of the SDD regimen and the concentrations of the constituent antimicrobials were measured. Enterobacteriaceae were not recovered from any of the tubes but six strains of Staphylococcus aureus, three Pseudomonas spp., three enterococci and four yeasts were isolated. Wide variations in the concentrations of all antibiotics were observed and in many cases they were below the MICs for the organisms isolated. In particular, tobramycin concentrations were uniformly less than the median MIC for the S. aureus isolates and this may account for the predominance of Gram-positive bacteria adherent to the ET tubes. Microbial biofilms attached to these tubes may have a role in the pathogenesis of nosocomial pneumonia in ICU patients.

Stability of ciprofloxacin in peritoneal dialysis solution.

The long-term stability of ciprofloxacin in dialysis fluid was studied. Ciprofloxacin was added to nine 2-L bags of dialysis solution containing 1.3% dextrose to yield a nominal concentration of 25 mg/L. Three bags each were stored at 4, 20, and 37 degrees C; three 20-mL samples were removed from each bag after 0, 0.5, 1, 2, 5, 7, 10, 14, 21, 28, and 42 days and analyzed in triplicate by high-performance liquid chromatography. Additional samples were removed from each bag on day 42 and analyzed by microbiological assay with Pseudomonas aeruginosa (nine samples tested for each storage temperature studied). The net percentage of change in ciprofloxacin concentration was 0.76% after storage at 4 degrees C, 1.02% after storage at 20 degrees C, and 0.75% after storage at 37 degrees C. Antimicrobial activity after storage at all three temperatures was confirmed by microbiological assay. Ciprofloxacin 25 mg/L was stable for 42 days when stored in dialysis fluid containing 1.36% dextrose at 4, 20, and 37 degrees C.

Cholecystokinin does not mediate glucose's cytoprotective effects.

Cold restraint stress produces acute gastric mucosal injury in association with altered gastric motility. Enteral nutrients prevent this injury in conjunction with inhibition of gastric emptying. Because cholecystokinin (CCK) is released by nutrients known to be cytoprotective and is thought to inhibit gastric emptying, we performed three experiments to see if CCK contributes to the gastric mucosal protection afforded by enteral nutrients. Our data show that enteral glucose protects the gastric mucosa and increases gastric volume, gastric luminal pH, and gastric mucin. Neither physiologic nor pharmacologic doses of CCK protected the mucosa. None of the other significant effects of glucose on gastric function during cold restraint were affected by exogenous CCK. Furthermore, antagonism of CCK receptors with L-364,718 did not have any independent effects, nor did it diminish the protection associated with enteral glucose. We conclude that enteral glucose protects the gastric mucosa from cold restraint injury in association with a number of potentially beneficial effects on gastric physiology, but none of the effects of glucose in this model appear to be mediated by CCK.

A pharmacokinetic and pharmacodynamic comparison of plain and enteric-coated prednisolone tablets.

1. Eight healthy volunteers and eight patients suffering from chronic obstructive pulmonary disease (COPD) received 30 mg prednisolone as plain (P) and enteric-coated tablets (EP) in a randomised, cross-over manner. Plasma prednisolone and cortisol and blood glucose were measured over 24 h. 2. Although absorption of prednisolone was considerably slower when administered as the enteric-coated form, peak plasma drug concentrations and total AUC (0,24 h) were equivalent for the two formulations. Malabsorption of prednisolone was not observed. 3. The administration of EP was associated with significantly less adrenal suppression in volunteers than P as judged by measurement of AUC (0,24 h) values for endogenous cortisol. However, this difference did not reach statistical significance in the patient group. 4. Plasma cortisol concentrations declined more slowly following administration of the enteric-coated form to both groups. The difference in time taken (median and range) to maximum suppression of cortisol was statistically significant (P less than 0.05) between P (2.5 h; 2-4 h) and EP (4 h; 3-12 h) preparations administered to volunteers. There was a similar significant difference (P less than 0.05) between P (2.5 h; 1-4 h) and EP (7 h; 2-12 h) in the patients. 5. Plasma cortisol concentrations were significantly lower at 24 h in patients receiving the enteric-coated product in association with higher terminal prednisolone concentrations. 6. Blood glucose concentrations increased over an 8 h period in both groups.(ABSTRACT TRUNCATED AT 250 WORDS)

Naloxone and restraint stress: effects on gastric mucosal injury and gastric function.

Contradictory findings have been reported from previous investigations of the effects of opiate receptor blockade on stress-related gastric injury. We performed two studies of the effects of naloxone on gastric mucosal injury and gastric function in cold restraint stress. In the first, naloxone had a linear, dose-dependent effect, reducing mucosal injury both parenterally and enterally. Gastric acidity was not related to the dose of naloxone, or to the amount or extent of mucosal injury. Gastric residual volume, by contrast, was related to both naloxone dose and gastric mucosal injury. In the second study we investigated the effect of gastric distention by enteral vehicle in the restraint model, and the interaction of the vehicle with the effects of naloxone. Periodic distention of the stomach with enteral vehicle did not alter the drug's effects on mucosal injury or on gastric residual volume. Because cold restraint stress produces harmful gastric contractions, naloxone is likely to be protective by altering gastric motility.

Enteral nutrients prevent stress ulceration and increase intragastric volume.

Tube feedings and intragastric glucose prevent stress ulceration by unknown mechanisms. We tested the hypothesis that glucose protects the gastric mucosa by direct repletion of glycogen stores. We compared the effects of enteral glucose with enteral lipids in the rat restraint model. The rats were given equal volumes of 0.9% saline, 20% lipids, or 25% glucose during a 4-h period of restraint stress. The effects of each treatment on gastric residual volume and luminal pH, as well as on stress lesion formation were measured. Both enteral nutrients significantly reduced the number of mucosal lesions compared to saline. In conjunction with their protective effect, both nutrients significantly increased both gastric residual volume and luminal pH. As stress-induced prolonged gastric contractions are related to mucosal injury in this model, the nutrient solutions may have been protective in part because they increased gastric volume and prevented mechanical trauma to the mucosa. We conclude that tube feedings do not prevent stress ulceration by glucose's repletion of local glycogen stores, as lipids and glucose were equally effective. Both increased intragastric volume and increased intraluminal pH associated with administration of enteral nutrients may contribute to their protection of the gastric mucosa from stress ulceration.

Pharmacokinetics of naloxone: an insight into the locus of effect on stress-ulceration.

The purpose of this study was to determine if the beneficial effect of naloxone on formation of acute gastric mucosal lesions was brought about via central or peripheral mechanisms by measuring blood concentrations of naloxone in rats during a 4-hr period of restraint stress. The study involved administration of naloxone to rats at doses of 5, 20 and 40 mg.kg-1.hr by either the intravenous or enteral routes. Blood samples were collected throughout the period of restraint and gastric stress-lesions were counted at the end of the experiments. Both routes of administration were equally effective in preventing stress-ulceration, with only rats receiving drug intravenously showing the presence of naloxone in blood samples. Inverse linear relationships existed between mean trough blood concentrations and lesions (P = .0003), as well as a linear correlation between area under the time-concentration curve and mean trough concentrations (P = .0001). Although our results show tight correlation between blood levels and effect on lesions in the group given drug intravenously, the effect must be on peripheral rather than central opiate receptors as no detectable blood levels were found when naloxone was given enterally.

Effect of quinolone antimicrobials on theophylline pharmacokinetics.

The purpose of the research was to ascertain the comparative differences of quinolone antibiotics on theophylline pharmacokinetics. Eight healthy male volunteers were randomly assigned to four treatments. Each was administered norfloxacin (NOR) 800 mg/d, ciprofloxacin (C) 1 g/d, nalidixic acid (NAL) 2 g/d and placebo (P) for 7 days. On the seventh day of each treatment, theophylline (5 mg/kg) iv was administered. The elimination half-life (T 1/2), total body clearance (CL) and volume of distribution at steady state (Vss) of theophylline were calculated using model-independent methods. ANOVA for repeated measures was used for data comparisons. The mean (SD) theophylline results were: CL l/kg/h--NOR .038 (.006), C .033 (.006), NAL .045 (.008), P .044 (.007); T 1/2 h--NOR 9.2 (1.8), C 10.6 (1.8), NAL 8.3 (1.8), P 7.5 (1.4). Theophylline Vss differences by treatment were not significant. NOR and C significantly decreased theophylline's clearance and the clearance change can be of clinical significance.

Naloxone exerts a dose-dependent gastric cytoprotective effect.

Previous work has shown that naloxone inhibits the ulcerogenic effects of indomethacin and stress, although the site, mechanism or dose are unknown. We investigated whether naloxone possessed gastric cytoprotective properties, generating a dose-response curve existed for both intragastric (IG) and intravenous (IV) administration. One hundred and two rats were subjected to a four hour period of restraint, with the last two hours at 4 degrees C. Naloxone was given hourly during restraint at doses of 0 (Control), 1, 5, 10, 20 mg/kg. After sacrifice, the residual gastric volume, and pH were measured and the number of mucosal lesions scored. The cytoprotection offered by naloxone was different from control (p = 0.0001), with the intravenous route having a greater effect (p = 0.038). While this protective effect did not correlate with changes in gastric acidity, it correlated with the dose of naloxone.

Stress ulcers: current understanding of pathogenesis and prophylaxis.

Stress-related gastrointestinal bleeding is known to occur in approximately 25 percent of untreated seriously ill patients, but with appropriate prophylaxis is largely preventable. Since the treatment of stress bleeding is generally unsatisfactory and has a high mortality, routine prophylaxis should be instituted for susceptible patients. Multiple mechanisms contribute to stress ulcer formation, the most important of which appear to be mucosal ischemia and the inability to control back-diffused hydrogen. Antacids and histamine2-blocking agents are presently the cornerstone of effective prophylaxis, but because they have been implicated as contributors to nosocomial pneumonias due to bacterial overgrowth in the stomach, investigation is ongoing into such alternative prophylactic agents as sucralfate and prostaglandins that do not alter the normal gastric acidity. This article presents a review of the literature on the development and prevention of stress ulcer disease.

Red cell pyrimidine 5' nucleotidase deficiency: determination of nucleotidase activity and nucleotide content using HPLC.

The hereditary deficiency of erythrocyte pyrimidine 5' nucleotidase has been investigated using an HPLC anion-exchange procedure designed to measure both red cell nucleotide content and enzymic activity. Red cell nucleotide profiles were determined using a low pH phosphate buffer salt-gradient system in 20 min, whereas a low pH buffer alone permitted the determination of enzymic activity with each of six different nucleotide substrates in less than 4 min. Both the red cell nucleotide profiles and the enzymic activity of haemolysates from two affected brothers and their children agreed well with previously published values. This unified approach should prove useful for detailed studies of deficiencies involving isoenzymes of pyrimidine nucleotidase.