A semi-supervised decision support system to facilitate antibiotic stewardship for urinary tract infections.

Urinary Tract Infections (UTIs) are among the most frequently occurring infections in the hospital. Urinalysis and urine culture are the main tools used for diagnosis. Whereas urinalysis is sufficiently sensitive for detecting UTI, it has a relatively low specificity, leading to unnecessary treatment with antibiotics and the risk of increasing antibiotic resistance. We performed an evaluation of the current diagnostic process with an expert-based label for UTI as outcome, retrospectively established using data from the Electronic Health Records. We found that the combination of urinalysis results with the Gram stain and other readily available parameters can be used effectively for predicting UTI. Based on the obtained information, we engineered a clinical decision support system (CDSS) using the reliable semi-supervised ensemble learning (RESSEL) method, and found it to be more accurate than urinalysis or the urine culture for prediction of UTI. The CDSS provides clinicians with this prediction within hours of ordering a culture and thereby enables them to hold off on prematurely prescribing antibiotics for UTI while awaiting the culture results.

Human monoclonal antibodies against Staphylococcus aureus surface antigens recognize in vitro and in vivo biofilm.

Implant-associated Staphylococcus aureus infections are difficult to treat because of biofilm formation. Bacteria in a biofilm are often insensitive to antibiotics and host immunity. Monoclonal antibodies (mAbs) could provide an alternative approach to improve the diagnosis and potential treatment of biofilm-related infections. Here, we show that mAbs targeting common surface components of S. aureus can recognize clinically relevant biofilm types. The mAbs were also shown to bind a collection of clinical isolates derived from different biofilm-associated infections (endocarditis, prosthetic joint, catheter). We identify two groups of antibodies: one group that uniquely binds S. aureus in biofilm state and one that recognizes S. aureus in both biofilm and planktonic state. Furthermore, we show that a mAb recognizing wall teichoic acid (clone 4497) specifically localizes to a subcutaneously implanted pre-colonized catheter in mice. In conclusion, we demonstrate the capacity of several human mAbs to detect S. aureus biofilms in vitro and in vivo.

Direct covalent attachment of silver nanoparticles on radical-rich plasma polymer films for antibacterial applications.

Prevention and treatment of biomaterial-associated infections (BAI) are imperative requirements for the effective and long-lasting function of orthopedic implants. Surface-functionalization of these materials with antibacterial agents, such as antibiotics, nanoparticles and peptides, is a promising approach to combat BAI. The well-known silver nanoparticles (AgNPs) in particular, although benefiting from strong and broad-range antibacterial efficiency, have been frequently associated with mammalian cell toxicity when physically adsorbed on biomaterials. The majority of irreversible immobilization techniques employed to fabricate AgNP-functionalized surfaces are based on wet-chemistry methods. However, these methods are typically substrate-dependent, complex, and time-consuming. Here we present a simple and dry strategy for the development of polymeric coatings used as platforms for the direct, linker-free covalent attachment of AgNPs onto solid surfaces using ion-assisted plasma polymerization. The resulting coating not only exhibits long-term antibiofilm efficiency against adherent Staphylococcus aureus (S. aureus), but also enhances osteoblast adhesion and proliferation. High resolution X-ray photoelectron spectroscopy (XPS), before and after sodium dodecyl sulfate (SDS) washing, confirms covalent bonding. The development of such silver-functionalized surfaces through a simple, plasma-based process holds great promise for the fabrication of implantable devices with improved tissue-implant integration and reduced biomaterial associated infections.

Interlaboratory comparison of sample preparation methods, database expansions, and cutoff values for identification of yeasts by matrix-assisted laser desorption ionization-time of flight mass spectrometry using a yeast test panel.

An interlaboratory study using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) to determine the identification of clinically important yeasts (n = 35) was performed at 11 clinical centers, one company, and one reference center using the Bruker Daltonics MALDI Biotyper system. The optimal cutoff for the MALDI-TOF MS score was investigated using receiver operating characteristic (ROC) curve analyses. The percentages of correct identifications were compared for different sample preparation methods and different databases. Logistic regression analysis was performed to analyze the association between the number of spectra in the database and the percentage of strains that were correctly identified. A total of 5,460 MALDI-TOF MS results were obtained. Using all results, the area under the ROC curve was 0.95 (95% confidence interval [CI], 0.94 to 0.96). With a sensitivity of 0.84 and a specificity of 0.97, a cutoff value of 1.7 was considered optimal. The overall percentage of correct identifications (formic acid-ethanol extraction method, score ≥ 1.7) was 61.5% when the commercial Bruker Daltonics database (BDAL) was used, and it increased to 86.8% by using an extended BDAL supplemented with a Centraalbureau voor Schimmelcultures (CBS)-KNAW Fungal Biodiversity Centre in-house database (BDAL+CBS in-house). A greater number of main spectra (MSP) in the database was associated with a higher percentage of correct identifications (odds ratio [OR], 1.10; 95% CI, 1.05 to 1.15; P < 0.01). The results from the direct transfer method ranged from 0% to 82.9% correct identifications, with the results of the top four centers ranging from 71.4% to 82.9% correct identifications. This study supports the use of a cutoff value of 1.7 for the identification of yeasts using MALDI-TOF MS. The inclusion of enough isolates of the same species in the database can enhance the proportion of correctly identified strains. Further optimization of the preparation methods, especially of the direct transfer method, may contribute to improved diagnosis of yeast-related infections.

Development of a real-time PCR for identification of brachyspira species in human colonic biopsies.

BACKGROUND: Brachyspira species are fastidious anaerobic microorganisms, that infect the colon of various animals. The genus contains both important pathogens of livestock as well as commensals. Two species are known to infect humans: B. aalborgi and B. pilosicoli. There is some evidence suggesting that the veterinary pathogenic B. pilosicoli is a potential zoonotic agent, however, since diagnosis in humans is based on histopathology of colon biopsies, species identification is not routinely performed in human materials. METHODS: The study population comprised 57 patients with microscopic evidence of Brachyspira infection and 26 patients with no histopathological evidence of Brachyspira infection. Concomitant faecal samples were available from three infected patients. Based on publically available 16S rDNA gene sequences of all Brachyspira species, species-specific primer sets were designed. DNA was extracted and tested by real-time PCR and 16S rDNA was sequenced. RESULTS: Sensitivity and specificity for identification of Brachyspira species in colon biopsies was 100% and 87.7% respectively. Sequencing revealed B. pilosicoli in 15.4% of patients, B. aalborgi in 76.9% and a third species, tentatively named "Brachyspira hominis", in 26.2%. Ten patients (12.3%) had a double and two (3.1%) a triple infection. The presence of Brachyspira pilosicoli was significantly associated with inflammatory changes in the colon-biopsy (p=0.028). CONCLUSIONS: This newly designed PCR allows for sub-differentiation of Brachyspira species in patient material and thus allows large-scaled surveillance studies to elucidate the pathogenicity of human Brachyspira infections. One-third of affected patients appeared to be infected with a novel species.

Clinical evaluation of a NASBA-based assay for detection of Candida spp. in blood and blood cultures.

The number of life-threatening opportunistic fungal infections has shown a dramatic increase. However, the diagnosis of candidemia remains difficult. Nucleic acid amplification assays may improve the detection rate and decrease the time needed for detection and identification of Candida spp. Whole blood samples of patients suspected of having candidemia were analyzed using Nucleic Acid Sequence-Based Amplification (NASBA). Furthermore, aliquots of blood cultures of the patients after 2 days of culturing were tested. Eleven data sets from ten patients in two hospitals were generated. None of the whole blood samples was positive in the NASBA assay. Eight samples were positive in the NASBA assay after two days of culturing, whereas only two additional positive samples were found after longer incubation periods. Thus, a two-day culture step is sufficient to greatly improve the sensitivity of the NASBA assay. The NASBA assay detected Candida RNA in three patients. In one patient, the yeast was not detected by automated blood culturing, in another patient the NASBA assay detected the infection two days earlier than the blood culture system.