Objective: To assess the clinical validity of polymerase chain reaction (PCR) based molecular methods in the microbiological diagnosis of culture negative infective endocarditis in a group of surgically treated patients. Design: Retrospective case-control study. Setting: Reference cardiovascular surgical centre. Patients and samples: 15 culture negative patients with infective endocarditis classified according to Duke criteria, with 17 heart valve samples; 13 age and sex matched control patients without infective endocarditis, with 13 valve samples. Interventions: Medical records were reviewed and clinical, demographic, and microbiological data collected, including results of molecular detection of bacteria and fungi from valve samples. The clinical validity of molecular diagnosis was assessed, along with the sensitivity and speed of the systems. Results: In the study group, 14 patients were PCR positive (93%). Organisms detected were streptococci (3), staphylococci (2), enterobacter (1), Tropheryma whippelii (1), Borrelia burgdorferi (1), Candida albicans (1), and Aspergillus species (2). Three cases were positive on universal bacterial detection but the pathogen could not be identified because of contaminating background. One case was negative. All but two positive cases showed clinical correlations. These two cases had no symptoms of infective endocarditis but there was agreement with the surgical findings. All control cases were PCR negative. Results were available within eight hours, and if sequencing was necessary, within 48 hours. Conclusions: PCR based molecular detection of pathogens in valve samples from surgically treated culture negative infective endocarditis patients is fast, sensitive, and reliable. The technology, combined with thorough validation and clinical interpretation, may be a promising tool for routine testing of infective endocarditis.

No abstract

Pichia fabianii, a yeast rarely causing human infections, was isolated from the blood of a patient with aortic valve endocarditis. The isolates were initially identified biochemically as Candida pelliculosa, but based on direct sequencing of the ITS2 region of rRNA, they were subsequently reidentified as P. fabianii. Antifungal therapy with fluconazole and later with voriconazole led to the development of resistant variants which had high MIC values to both antifungals. Strong biofilm formation by this yeast could also have played a role in the development of its resistance and allowed for its persistence on the infected valve during antifungal therapy. To our knowledge, this is the first published case of endocarditis and the fourth human infection caused by this yeast species.

No abstract

Prosthetic joint infection (PJI) diagnosis includes several classes of verification. Among them, only a few have a stronger independent value, namely intraarticular purulence and communicating fistulas. Other diagnostic methods require careful test combinations, analysis, and interpretation. Molecular based techniques using the polymerase chain reaction (PCR) seem to be a promising PJI diagnostic modality due to its excellent sensitivity, specificity, positive predictive value, and speed. Most of the recent reviewers are in agreement that molecular diagnosis has enough potential for future application in orthopaedics even if there are only a few heterogeneous studies fully supporting this concept. Conversely, at least one study has been published with significantly worse results (sensitivity and specificity less than 0.75). The lack of supporting evidence in the published studies may be closely related to varying PCR laboratory procedures, inappropriate reference standards, and other methodological shortcomings among research centers. It is not yet justifiable to firmly include molecular methods into the present PJI diagnostic schemes. The orthopaedic community must await the results of well-organized ongoing studies before considering inclusion of molecular diagnostics as a PJI diagnostic method. The aim of this paper was to make a survey of current PJI molecular diagnostic techniques in orthopaedics.

Our results suggest that PCR-APLP-RFLP assays can be useful in the diagnosis of fungal infections in immunocompromised patients.

During the years 1996-2000, a total of 2398 Ixodes ricinus ticks were collected in three areas of southern Moravia and eastern Bohemia, Czech Republic, and examined by dark-field microscopy for the presence of spirochetes. The prevalence of B. burgdorferi sensu lato in Ixodes ricinus ticks varied and depended upon the year. In 1996, the prevalence 6.8% was observed, during 1997 and 1998, it increased to 8.4% and 12.3%, respectively. The lowest prevalence was observed in 1999 (3.6%), and in 2000 it increased to 4.0%. The mean rate of infection was 6.5%, and the proportions of infected ticks were 12.2% in 263 male ticks, 8.3% in 289 female ticks, 6.0% in 1621 nymphs, and 1.3% in 225 larvae. From the total of 156 highly infected ticks (>100 spirochetes per sample) transferred into BSK-H medium for isolation attempts, 13 isolates were obtained. PCR-RFLP and electrophoresis (SDS-PAGE) were used for the identification and characterization of Borrelia strains. Ten tick isolates were identified as Borrelia afzelii, and the other three isolates were found to be Borrelia garinii. The results indicate the epidemiological importance of B. afzelii and B. garinii in central Europe, and emphasize the role of I. ricinus in the ecology of B. burgdorferi and in the epidemiology and epizootiology of Lyme disease.

Isolates from the "farm to fork" samples (182 isolates from 2779 samples) were examined genotypically (icaAB genes) and phenotypically (in vitro biofilm formation, typical growth on Congo red agar; CRA) with the aim to assess the risk of penetration of virulent strains of Staphylococcus epidermidis into the food chain. The contamination of meat and milk products was significantly higher in comparison with raw materials. Contamination of contact surfaces in the meat-processing plants was significantly lower than that of contact surfaces in the dairy plants. The ica genes (which precondition the biofilm formation) were concurrently detected in 20 isolates that also showed a typical growth on CRA. Two ica operon-negative isolates produced biofilm in vitro but perhaps by an ica-independent mechanism. The surfaces in the dairy plants and the milk products were more frequently contaminated with ica operon-positive strains (2.3 and 1.2 % samples) than the other sample types (0-0.6 % samples).