High-level disinfection is employed throughout the health services in the disinfection of medical equipment, such as endoscopes, to prevent patient-to-patient infections. The likelihood of an endoscope transmitted infection occurring is rare, providing strict guidelines are followed for effective decontamination between procedures. Endoscopes are subjected to rigorous cleaning and high-level disinfection within washer-disinfectors. However, poor decontamination protocols and inappropriate use of disinfectants can lead to can lead to incomplete disinfection and resistance. A number of bacterial strains were isolated from endoscope washer-disinfectors on several occasions. The efficacy of high-level disinfectants (chlorine dioxide, peracetic acid and hydrogen peroxide-based) against these isolates was measured using standard efficacy tests. Resistance mechanisms involved in bacterial survival following biocide exposure were investigated using scanning and transmission electron-microscopy for gross-morphology changes, measurements of expression of detoxifying enzyme and RT-PCR for resistance genes expression, while the role of extracellular polysaccharide in decreasing biocide efficacy, was studied. Two bacterial isolates (Bacillus subtilis and Micrococcus luteus) were shown to have a high resistance to chlorine dioxide. Electron microscopy showed significant differences between isolates and reference strains. The B. subtilis isolate produced large quantities of extracellular polysaccharide, which may be interfering with biocide activity. Genes for catalase and superoxide dismutase were present in B. subtilis and enzyme activity varied between isolates and reference strains, indicating a potential involvement in resistance mechanisms, however the extent remains unclear. It was found that the isolate extracellular polysaccharide was not involved in conferring resistance to oxidising agents. This study demonstrated that bacteria can survive high-level disinfection with oxidising agents and that mechanisms conferring resistance are complex but might not be linked to impaired biocide penetration. Furthermore, the findings of this work show that surveillance programmes are essential for monitoring the incidence of biocide resistant isolates in the healthcare environment.

In cystic fibrosis (CF), bacteria of the Burkholderia cepacia complex (Bcc) can induce a fulminant inflammation with pneumonitis and sepsis. Lipopolysaccharide (LPS) may be an important virulence factor associated with the increased morbidity and mortality seen in Bcc infection but little is known about the molecular pathogenesis of Bcc LPS. In this study, the inflammatory response to highly purified LPS from different Bcc clinical isolates and the cellular signalling pathways employed was investigated. It has been suggested that the large inflammatory response from BcLPS may be due to contaminants in the LPS preparation. Phenol-chloroform petroleum ether (PCP) purified BcLPS preparations were compared to more highly purified, enzyme treated preparations of BcLPS. There were no significant differences in the levels of IL-6 and TNF-a induced from monocytes (MM6) and in levels of IL-8 from epithelial cells (A549), which indicates that there were no contaminants present that could cause an inflammatory response from these cells. The inflammatory response elicited by LPS from different Bcc species that were tested was seen to be varied. LPS from different clinical isolates of the same clonal ET12 strain of Burkholderia cenocepacia were found to induce a varied inflammatory response. Some isolate's LPS produced as much cytokine as prototypical Escherichia coli LPS and all Bcc isolates tested with the exception of environmental samples produced higher levels of inflammatory cytokine than LPS from the CF pathogen Pseudomonas aeruginosa. It was shown that passaging over time under laboratory conditions was not responsible for this variation. Bcc LPS samples were tested in their ability to prime MM6 cells for respiratory burst, samples that had previously produced high levels of cytokine from direct stimulation of MM6 cells were found to not be the most efficient primers of respiratory burst. It was found that the inflammatory response from Bcc LPS stimulated monocytes was separate and in some cases opposite to the ability to prime for respiratory burst. Inhibition experiments were used to investigate the Toll-like receptors and associated adaptor molecules and pathways utilized when monocytes were stimulated by Bcc LPS. The use of anti-CD14, anti-TLR4 and anti-TLR2 antibodies in Bcc LPS stimulated monocytes showed that all Bcc LPS samples tested signalled via CD 14 and TLR4 and not via TLR2. Inhibition of MyD88 using an inhibitor peptide proved that all but one sample required MyD88 to signal. LPS from clinical isolate of Burkholderia multivorans was found to activate the inflammatory response via MyD88-independent pathways. Using MAP kinase inhibitors to test for reduction of cytokine response from stimulated MM6 cells and direct analysis of activation through western blotting, LPS from all clinical Bcc isolates were seen to activate all three major MAPKs p42/44, p38 and JNK. Degradation of the NF-kB inhibitory protein IicB-a was tested using anti-lKB-a antibodies and activation of the transcription factor NF-kB was tested using an electrophoretic mobility shift assay (EMSA). IicB-a was degraded and NF-kB was activated in all the BcLPS samples tested. This study suggests that LPS alone from clinical isolates of Bcc is major virulence factor in CF that it can cause a massive inflammatory response from cells, and that the LPS induced signalling cascade is via classical TLR4-mediated signalling pathways similar to highly inflammatory LPS purified from E.coli.

Chronic skin wounds (CW) represent a significant world health problem including reservoirs of multi-drug resistant bacterial biofilms. The precise role of bacteria in the aetiology of chronic inflammation and non-healing remains unclear. This study characterised MRSA from human CW and investigated how they inhibit wound healing, modulate immunological responses and resist treatment in comparison to control MRSA from asymptomatic nasal carriers (NC). Routine cultural analysis of 150 chronic wounds revealed 50% were colonised with <italic>S. aureus</italic>, of which 22.6% were MRSA. Multi-locus sequence typing identified two new sequence types and demonstrated that wound MRSA represented two clonal complexes (22 and 30) with almost 90% identified as hospital-acquired EMRSA-15. MRSA isolated from CW and NC were characterised for virulence factor (VF) expression and modulation of the innate immune system. Presence and expression of MRSA VFs indicated an association with sequence type. Greater expression of colonisation- (cna) and degradation- associated (hysA) VFs was evident in the wound MRSA, suggesting that modulation of virulence is important for non-healing. The ability of conventional biocides (iodine, silver, and potassium permanganate) to treat chronic wound bacteria, using the carrier-test method, revealed iodine as the only effective biocide. In vitro stimulation of the ability of the MRSA to induce innate immunity showed that CW-MRSA exhibited decreased TLR, cytokine and complement responses compared with NC-MRSA (IL-8, TNFa, complement activation PO.05). Moreover, biofilm- induced reductions in immunogenicity were observed compared with planktonic growth in monocyte and complement assays (PO.05). Scratch wound assays indicated that MRSA failed to inhibit keratinocyte migration (P>0.05), although bacterial growth conditions (biofilm vs. planktonic) significantly affected the observed cellular migration (PO.05). Virulence factor production and ability to modulate/evade the host innate immune response are important potential mechanisms by which MRSA are able to colonise chronic wounds. These studies provide important new insights into the role of MRSA in delayed dermal healing.

The aim of this study was to compare the abilities of various urease-positive species to encrust and block catheters with crystalline biofllm. Experiments were performed in laboratory models of the catheterised bladder infected with a range of urease- producing species. The results of these experiments allowed the classification of the bacteria into three groups: rapid encrusters, slow encrusters and non-encrusters. Rapid encrusters (<italic>Pr. mirabilis, Proteus vulgaris</italic> and <italic>Providencia rettgeri</italic>) were able to raise the urinary pH to 8 - 9 and cause catheter blockage within 37 h. Slow encrusters (<italic>Morganella morganii, Staphylococcus aureus</italic> and <italic> Staphylococcus saprophytics</italic>) were able to raise the urinary pH moderately (from 6.1 to 6.89 - 7.39 over 96 h) and cause the formation of some encrustation on the catheters. Non- encrusters {<italic>Providencia stuartii, Pseudomonas aeruginosa, Klebsiella pneumoniae, Klebsiella oxytoca, Serratia marcescens, Enterobacter cloacae, Citrobacter freundii</italic> and <italic>Citrobacter koseri</italic>) were not able to raise the mean urinary pH above mean levels of 6.45 and did not form crystalline biofilm. <italic>Pr. mirabilis, Pr. vulgaris </italic> and <italic>Pv. rettgeri</italic> were also capable of rapidly encrusting silver-hydrogel coated latex catheters (Bard I.C. catheter) and nitrofurazone impregnated silicone catheters (Rochester NF catheter). There were no significant differences between the times these organisms took to block these catheters compared to all-silicone control catheters. The antimicrobial catheters also had no effect on the urinary pH generated by these organisms. The insensitivity of the three encrusting species to nitrofurazone (MICs 32- 128 ug/ml) is clearly a major factor in the failure of these catheters to prevent encrustation. The results of experiments in which the balloons of all-silicone catheters were inflated with solutions of triclosan (3 mg/ml in 0.1 M sodium carbonate) confirmed previous observations that catheter encrustation by <italic>Pr. mirabilis</italic> was prevented by this strategy. It also proved effective against <italic>Pr. vulgaris</italic>. In both cases, in contrast to the controls, the numbers of viable cells recovered from the residual urine fell steeply within 24 h, the pH of the urine dropped below its nucleation pH (pH 6.5) and the catheters drained freely for the seven day experimental period. The effect of triclosan on encrustation by Pv. rettgeri was minimal however, with no significant difference between blockage times or urinary pHs in the test and control models. While the Proteus sp. had MICs of triclosan of < 0.2 ug/ml, the value for <italic>Pv. rettgeri</italic> was 64 ug/ml. Inflating catheter balloons with a solution which generated nitric oxide proved ineffective as a means of controlling catheter encrustation. Previous studies have shown that a simple cellulose acetate / bromothymol blue sensor is capable of signalling infection by <italic>Pr. mirabilis</italic> and the early stages of catheter encrustation. Placed in the drainage bag it can give early warnings to patients, carers and nurses that catheters need to be replaced. While the use of the sensor in this way could avoid the clinical crises induced by catheter blockage, it would be of more value if an effective strategy to inhibit encrustation could be deployed when the problem is signalled. In the present study it was demonstrated that strips of the sensor polymer placed in the drainage bags changed from yellow to blue signalling the rise in urinary pH induced by infection with <italic>Pr. mirabilis, Pr. vulgaris</italic> or <italic>Pv. rettgeri </italic>. Electron microscopy confirmed that encrustation had started on the catheters at the times the sensors turned blue. Triclosan (3 mg/ml) introduced into the catheter balloons when the signal was observed was found to halt the development of the Proteus crystalline biofilms. It was concluded that an integrated sensor / modulator strategy was feasible for the control of encrustation by these species. (Abstract shortened by UMI.).

Infection by <italic>Proteus mirabilis</italic> can seriously complicate the care of patients undergoing long-term indwelling bladder catheterisation. The urease-producing bacteria colonize the catheter surfaces forming extensive biofilm communities and are capable of generating ammonia from urea and elevating the pH of the urine and biofilm. Under these conditions crystals of calcium and magnesium phosphates form in the urine and within the bacterial biofilm on the indwelling device leading to its encrustation and blockage. Urine can leak around the outside of the blocked catheter and patients become incontinent. Alternatively, urine is retained within the bladder, causing painful distension of the bladder. Reflux of infected urine to the kidneys can lead to serious symptomatic episodes such as pyelonephritis, septicaemia and endotoxic shock. All available types of indwelling catheter are vulnerable to this problem and currently there are no effective procedures available for its control. While the basic mechanism has been established for catheter encrustation we still need to know more about some of the fundamental aspects of the process. Little is known about the early events and the precise mechanisms which <italic> P. mirabilis</italic> uses to colonize catheter surfaces. The factors that control the rate at which crystalline biofilm forms on the catheters are also unknown. The aims of this study were to establish the sequence of events in the early stages of crystalline <italic>P. mirabilis</italic> biofilm formation on the range of currently available catheters for use with patients to determine the role of Mannose-Resistant Proteus-hkc fimbriae (MR/P fimbriae) in <italic> P. mirabilis</italic> crystalline biofilm formation on catheters to investigate how the pH at which calcium and magnesium phosphates precipitate from urine, the nucleation pH (pHn) can be manipulated and to determine the effect of this parameter on the rate of catheter encrustation. Using a laboratory model of the catheterised bladder, scanning electron microscopy and X-ray microanalysis, the initial stages of <italic>P. mirabilis</italic> crystalline biofilm development was observed on catheter surfaces. All-silicone, silicone-coated latex, hydrogel-coated latex and hydrogel/silver-coated latex catheters rapidly acquired a microcrystalline 'foundation layer' comprised predominantly of calcium phosphate, upon which, <italic>P. mirabilis</italic> crystalline biofilm subsequently developed. A similar 'foundation layer' was observed on the encrusted surfaces of hydrogel/silver-coated catheters removed from long-term catheterised patients. The catheters impregnated with nitrofurazone briefly delayed the onset of crystalline biofilm formation, while all-silicone and hydrogel-coated latex catheters inflated with triclosan (3 mg/ml in Na2C03) were able to maintain acidic urine pH and prevent crystalline biofilm development for the 7 day experimental period. There is evidence that MR/P fimbriae are involved in initiating infection in non-catheterised urinary tracts. The role of these adhesins in crystalline biofilm formation on indwelling catheters however, has not been investigated. Using bladder models infected with a wild type <italic> P. mirabilis</italic> strain able to express MR/P fimbriae and its derived MR/P-negative mutant, time to catheter blockage experiments and scanning electron microscopy revealed that MR/P fimbriae were not essential for <italic>P. mirabilis </italic> colonization of catheter surfaces or the development of crystalline <italic> P. mirabilis</italic> biofilm. Although the wild type and mutant strain initiated biofilm formation in different ways both rapidly blocked all-silicone catheters with crystalline material. The overriding factor in catheter blockage was the generation of alkaline urine, raising the pH above that at which crystalline formations develop. Previously it has been demonstrated that the pHn of human urine can be elevated by dilution and by increasing its citrate content. In the present study the effect of dilution and adding citrate on the pHn of artificial was assessed. Furthermore, the effect on the rate of encrustation on all-silicone catheters was examined in laboratory models supplied with these urines and infected with urease-positive <italic>P. mirabilis, Providencia rettgeri</italic> and <italic>Proteus vulgaris.</italic> The pHn of urine could be elevated from pH 6.7 in neat urine to pH 8.4 in urine diluted to 1:6. When neat, 1:1, 1:2 and 1:3 diluted urines were supplied to bladder models significant increases in catheter lifespan were recorded at each ascending dilution. Increasing the citrate content of the 1:1 diluted urine from 0 to 3.0 g/L citrate elevated the pHn from pH 7 to pH 9.1. Scanning electron microscopy of catheter sections revealed crystalline material in the biofilms could be virtually eliminated for at least 7 days in models supplied with urine with pHns of >pH 8.5. Time to catheter blockage experiments showed the rate of catheter encrustation became significantly reduced as the pHn of urine increased. Catheters in models supplied with urine containing citrate concentrations of 1.5 mg/ml (pHn >8.4) or more drained freely for the whole 7-day experimental period.

Hand hygiene is the most important measure for reducing healthcare associated infections. Previous in-vitro testing of alcohol hand rubs AHR has not reflected conditions found in practice therefore little is known about their true efficacy. This study aimed to discover how effective AHRs are when used in local intensive therapy Units ITU against Staphylococcus aureus clinical isolates and their mode of action. The AHRs were unable to achieve significant bactericidal effect 4 log reduction against S. aureus within the time healthcare workers HCW took to rub AHR into their hands, particularly when tested against bacterial cells on the surface of excised skin. The AHRs had no residual effect and the mechanical action of hand rubbing only increased bacterial cell death by approximately 1 log. The AHRs demonstrated varying efficacies against the isolates and variation in susceptibility to the AHRs was also observed among the isolates. AHR damaged the cytoplasmic membrane, but this was not the cause of the bactericidal effect, nor a reason for the differences in susceptibility observed among the strains. The cell wall and/or outer layers of the bacterial cell were found to most likely be the main target of AHR, most likely due to protein denaturation, although this study was unable to confirm this hypothesis. Virulence gene expression, cell surface charge and cell surface hydrophobicity was not affected by exposure of bacterial cells to AHR and were not responsible for varying AHR susceptibility among the isolates. This study determined the quality of AHR application amongst HCWs in the ITU and the potential cross-contamination resulting from failure to perform hand hygiene before and after certain activities. It can be concluded from this study that AHRs are unlikely to kill all microbial flora on HCWs hands. Recommendations have been made to improve hand hygiene practices and the efficacy of AHRs used in ITUs.

Chronic wounds cause substantial morbidity and healthcare costs and prevalence is rising as the population ages and diabetes increases. Microbes are ubiquitous in chronic wounds, with Staphylococcus aureus and Pseudomonas aeruginosa commonplace. Antibiotic resistance is also widespread and increasing. Patients with chronic wounds are exposed to many antibiotic resistance risk factors. This study investigated antibiotic consumption by patients with chronic wounds and the prevalence of and risk factors for antibiotic resistant organisms in such wounds. Finally, the impact of resistance on the cost of treatment was investigated. Antibiotic consumption by patients with chronic wounds treated in primary care was significantly higher than matched patients without chronic wounds. This included greater quantities of flucloxacillin, co-amoxiclav, metronidazole, and ciprofloxacin. The prevalence of antibiotic resistant organisms in chronic wounds of patients attending a specialist wound-healing clinic was investigated. No patients carried vancomycin- resistant enterococci in their wounds. The prevalence of methicillin-resistant S. aureus (MRSA) was 10%. No wound characteristics were associated with MRSA. Carriage was associated with previous MRSA and 'other' systemic antibiotics. The prevalence of ciprofloxacin-resistant P. aeruginosa was 11%. Exploratory analysis identified previous antibiotics (specifically ciprofloxacin, 'other' topical antimicrobials and 'other' systemic antibiotics) and wound aetiology as risk factors. Healing wounds were less likely to carry ciprofloxacin-resistant P. aeruginosa. Treatment costs for venous leg ulcers were explored using Markov models: one year's treatment, following presentation, cost &pound;1008. Antibiotic resistance prevalence had little impact on cost. The frequency of nursing visits (for healed and active ulcers), cost of hospital appointments and cost of nurses had the greatest impact. In summary, antibiotics are commonly used in primary care management of chronic wounds. However ciprofloxacin and 'other' systemic antibiotics may be associated with carriage of antibiotic resistant organisms. The impact of resistance on treatment costs of venous ulcers is small, provided effective alternatives are available.

Around 50% of patients enduring long-term catheterisation experience encrustation and blockage of their catheters. This problem stems from infection by urease producing bacterial species, in particular Proteus mirabilis. The urease enzyme hydrolyses urea to carbon dioxide and ammonia which elevates the urinary pH. Under these alkaline conditions crystals of calcium and magnesium phosphates form and the crystalline bacterial biofilm that develops on the catheter can eventually block the flow of urine through its lumen. Catheter blockage in this way can induce complications that put the health of the patient at serious risk. Currently there are no effective methods for controlling this process and little is known about the bacterial or host factors that might modulate the rate of catheter encrustation. The extent to which catheter biofilms contain potentially dangerous levels of endotoxin is also unknown. In view of the lack of information relating to these issues the objectives of this study were to: (a) gain an insight into the complexity of the urinary flora of patients undergoing long-term catheterisation (b) examine the bacterial composition of catheter biofilms for evidence of antagonisms between Pr. mirabilis and other species (c) test the effects of other uropathogens on the ability of Pr. mirabilis to produce catheter encrustations in laboratory models of the catheterised bladder (d) examine the hypothesis that coaggregation between Pr. mirabilis and other species is involved in the formation of crystalline catheter biofilms and (e) determine whether endotoxin can be found in catheter biofilms from patients undergoing long-term catheterisation. Over a six-week period urine samples were analysed from five patients undergoing long-term catheterisation. The urinary flora was both polymicrobial and dynamic, commonly containing at least four bacterial species. The pH of the urine varied from week to week. The presence of Pr. mirabilis was always associated with alkaline urine (mean pH 8.66). The presence of other urease producing species such as Pseudomonas aeruginosa and Morganella morganii were not associated with highly alkaline urine. In the cases of the four patients who did not suffer from catheter blockages in the study period, the nucleation pH (pHn) of their urine at week six was above the pH of their voided urine (pHv). The only patient in which the pHn was below the pHv had a stable Pr. mirabilis infection and had two catheters block during the study period. A significant negative correlation was found between the urinary concentrations of calcium and magnesium and the nucleation pH value. Strategies to decrease the concentrations of these divalent cations will act to increase the nucleation pH and reduce the rate of crystal accumulation and mineralised bacterial biofilm development. To control catheter encrustation it will be essential to prevent the ability of Pr. mirabilis to elevate the pH of the urine above its nucleation pH. Analysis of the data on 106 catheter biofilm communities from long-term hospital and community-dwelling catheterised individuals revealed that the overall incidence of Pr. mirabilis was 30.19%. Particularly when species such as Klebsiella pneumoniae were recovered from catheters, the percentage incidence of Pr. mirabilis was above this figure. In contrast, when species such as Escherichia coli, Morg. morganii or Enterobacter cloacae were present on a catheter, Pr. mirabilis was rarely or never found. An experimental approach, using laboratory models of the catheterised bladder, was used to investigate the interactions of Pr. mirabilis with the test organisms Et. cloacae, Morg. morganii, Kl. pneumoniae, E. coli, and Ps. aeruginosa in more detail. Experiments in laboratory models showed that super-infection of Pr. mirabilis after 24 h growth of one of each of the test species had little or no effect on the ability of Pr. mirabilis to encrust and block catheters. However, growth of Et. cloacae, Morg. morganii, Kl. pneumoniae, or E. coli for 72 h prior to Pr. mirabilis super-infection significantly delayed catheter blockage. When Pr. mirabilis was inoculated into models 72 h after Et. cloacae for example, the mean time to blockage was extended from 28.74 h to 60.73 h (P &lt; 0.01). In all cases however, Pr. mirabilis was eventually able to generate alkaline urine, induce crystal formation, and block the catheters. Reconstituting a four-member bacterial community from this patient significantly slowed the rise in urinary pH and postponed blockage compared to models infected with the Pr. mirabilis alone. Biofilms on sections of catheters received from patients were found to contain endotoxin levels ranging from 282.8 to 917.2 ng/4 cm length of catheter. The results from this study suggest that antagonistic interactions between Pr. mirabilis and other urinary tract organisms do exist. (Abstract shortened by UMI.)

Clostridium difficile infection (CDI) is the most common type of infectious nosocomial diarrhoea. Overwhelming evidence indicate that the most important risk factors are prior antibiotic use and elderly patients. The severity of the disease varies from asymptomatic carrier to mild diarrhoea to colitis (AAC) and life threatening pseudomembranous colitis (PMC). Because little is known about C. difficile and CDI in Kuwait, this study was undertaken to determine the nosocomial acquisition of C. difficile by new patients admitted to the intensive care units (ICU) of 4 teaching hospitals in Kuwait between February 2001 and January 2002 (first part) and January 2003 to December 2005 (second part) and evaluate cytotoxin (toxin B) production by clinical isolates upon exposure to minimum inhibitory concentrations (MICs) and sub-MICs of certain antibiotics. The first part of the study was accomplished by serially culturing the stool specimens of 922 newly admitted patients to the ICUs screening their stools for toxins A/B and screening their immediate environment for C. difficile. The isolates were typed by the PCR ribotyping technique developed in the Anaerobe Reference Unit, Cardiff. The effects of various concentrations of antibiotics that could predispose to CDI and those used for its therapy on the production of cell-bound and cell-free toxin B produced by C difficile was investigated by experiments using cell cultures of the Vero cell line. Prevalence, epidemiology and risk factors of CDI in Kuwait hospitals was investigated during second part of the study by culturing patients' stool specimens, ribotyping the isolates and detection of toxin A/B in stool samples. The susceptibility of all isolates was assessed by MIC determination to 16 antibiotics using the E test method. During the first part of the study, 95 (10.3%) out of 922 patients with negative cultures initially on the day of admission acquired C difficile during their hospitalisation at various time intervals. Of these, 65 (68%) remained symptom-free while 30 (32%) were symptomatic 2 patients had PMC, 4 AAC and 24 AAD. C. difficile toxin A/B was present in 28 (93%) of 30 symptomatic patients but in only 7 (10.8%) of 65 symptom-free patients. The hospital environments occupied by symptomatic patients as well as those occupied by asymptomatic patients were contaminated by C. difficile. The 95 isolates from patients belonged to a total of 32 different ribotypes. Ribotypes 097 and 078 were responsible for >40% of C. difficile infections in Kuwait ICUs. There was a heterogeneous relationship between antibiotic exposure and intra- and extra cellular toxin production by the toxigenic strains. Clinical strains of C. difficile when exposed to MIC and sub-inhibitory concentrations of certain antibiotics produced high level of cytotoxin. Ampicillin and clindamycin were the most potent inducers of cytotoxin followed by metronidazole and vancomycin. Cefotaxime induced the least amount of the cytotoxin activity. During the second part of the study, 73 (10.5%) out of 697 met the diagnosis of CDI. Out of these 73, 56 (76.7%) were hospital-acquired and 17 (23.3%) were from outpatient clinics. Thus, the prevalence of hospital-acquired CDI was 8% over the study period. The prevalence of hospital-acquired CDI in 2003, 2004 and 2005 were, 9.7%, 7.8% and 7.2%, respectively. Our data showed that 42.9% of the CDI patients were above 60 years out of which over 79% were aged 71 years and above. Patients with CDI were more likely than the controls to have been exposed to immunosuppressive drugs and feeding via naso-gastric tube. The most common ribotypes isolated during the second part of the study were 002 and 001. The later was isolated only from one environmental sample in the first part of the study. PCR-ribotype 027 was not isolated during 2003-2005 study. None of our 151 C. difficile isolates were resistant to amoxicillin-clavulanic acid, ampicillin, linezolid, metronidazole, piperacillin-tazobactam, teicoplanin or vancomycin. Resistance to penicillin and meropenem among the clinical isolates increased from 2.4 to 16.4% and 4.8 to 21.4%, respectively while resistance to imipenem (another carbapenem) was extremely high in both studies.

Staphylococcus aureus is an extra-ordinarily versatile pathogen causing a wide spectrum of infections. The aims of this study are to analyze 10 clinical isolates of S. aureus from the UK by Multi Locus Sequence Typing (MLST) and determining their PVL-type variants. In addition to that, to study the effect of several antibiotics at sub inhibitory concentrations on a number of virulence factors at mRNA using quantitative PCR and protein levels using proteomic methods. Western blotting was used to study differential expression of Spa at protein levels. Data showed that the 10 clinical isolates belong to seven clonal complexes (CCs), which are CC1, CC5, CC8, CC22, CC30, CC88, and CC121. Genetic variation within lukSF-PV gene showed that three of these isolates were belong to the same PVL type variant of CA-MRSA USA300 strain, R variant. From which, two isolates were found to belong to the same CC of USA300, CC8. The remaining 7 isolates were found to belong to H variant. Data presented here showed that the sub-MIC levels of both cell wall inhibitors reduced lukSF-PV and spa steady-state mRNA levels when cells were grown in the presence of these antibiotics for one hour. However, after 5 hrs post antibiotic addition of these two antibiotics, vancomycin remained depressed lukSF-PV and spa steady-state mRNA levels as well as at protein levels, but oxacillin increased spa and lukSF-PV mRNA levels, as well as Spa at protein levels. Protein synthesis inhibitors clindamycin and linezolid were both caused an increase of lukSF-PV mRNA levels, but they both decreased spa mRNA levels, when cultures grown in the presence of these antibiotics for one hour. However, when cultures grown with these antibiotics for 5 hrs, clindamycin remained to increase lukSF-PV and decrease spa mRNA levels and protein levels, but linezolid decreased both virulence factors at mRNA and protein levels. The data showed in this study confirmed that growing S. aureus in the presence of oxacillin induce toxin expression and might enhance the virulence of this bacterium, therefore using these antibiotics to treat S. aureus infections may contribute to worse outcomes. These data also confirmed that linezolid and vancomycin, are both important selections of antimicrobial agents to treat serious infections caused by the bacterium.