Neonatal intestinal colonization with extended-spectrum ß-lactamase-producing Enterobacteriaceae-a 5-year follow-up study.

OBJECTIVES: To analyse Klebsiella pneumoniae (KP) isolates from an outbreak of extended-spectrum ß-lactamase (ESBL)-producing KP and Escherichia coli (EC) among infants admitted to neonatal intensive care units and to determine the duration of the intestinal colonization. METHODS: We performed a prospective cohort study of intestinal ESBL-KP/ESBL-EC colonized neonates after a 5-month outbreak in two neonatal intensive care units. Whole genome sequencing, multilocus sequence typing, core genome multilocus sequence typing, pulsed-field electrophoresis and PCR for blaCTX-M were performed on the first isolates. Stool cultures were performed every second month after discharge until 2 years after discharge and at 5 years of age. The last positive samples were analysed with pulsed-field gel electrophoresis and PCR for blaCTX-M. The intestinal relative dominance of ESBL-producing Enterobacteriaceae was determined. RESULTS: Thirteen of 17 patients colonized with ESBL-KP/ESBL-EC survived. Isolates from 16 of 17 patients were available for analysis and featured the same strain type of ESBL-KP: sequence type 101. The strain had capsule type K29 and harboured blaCTX-M-15. The virulence genes irp1, irp2, iutA, kfu and mrk were detected in all isolates. The median length of colonization was 12.5 months (range, 5-68 months). After 2 years, two of 13 patients were carriers of ESBL-KP and one of 13 of ESBL-EC. At 5 years of age, one neonate was colonized with ESBL-EC. No infant experienced an ESBL-KP/EC-infection during follow-up. CONCLUSIONS: Two years after discharge, almost one fourth of the study participants were ESBL/KP-EC carriers. ESBL-KP sequence type 101 persisted in two of 13 children for 23 to 26 months. One patient was colonized with ESBL-EC at age 5 years.

Alarming spread of vancomycin resistant enterococci in Sweden since 2007.

The total number of persons infected or colonised with vancomycin-resistant enterococci mandatorily reported to the Swedish Institute for Infectious Disease Control increased dramatically during 2007 and 2008. During a period of twenty months from 1 July 2007 to 28 February 2009, a total of 760 cases were reported compared with 194 cases reported during the entire period from 2000 to 2006. This rise was mainly attributed to a wide dissemination of vancomycin resistant enterococci which started in a number of hospitals in Stockholm in the autumn of 2007 and was followed by dissemination in various healthcare facilities (hospitals and homes for the elderly) in a further two Swedish counties in 2008. The majority of the cases (97%) were acquired in Sweden and among these, healthcare-acquired E. faecium vanB dominated (n=634). The majority of these isolates had identical or closely related pulsed-field gel electrophoresis patterns indicating clonal dissemination in the affected counties. The median minimum inhibitory concentration of vancomycin was 32 mg/L (ranging from 4 to >128 mg/L) and of teichoplanin 0.12 mg/L (ranging from 0.06 to 0.25 mg/L). Particular emphasis was placed on countermeasures such as screening, contact tracing, cleaning procedures, education in accurate use of infection control practices as well as increasing awareness of hygiene among patients and visitors. With these measures the dissemination rate decreased substantially, but new infections with the E. faecium vanB strain were still detected.

Hospital outbreak control requires joint efforts from hospital management, microbiology and infection control.

An outbreak of multidrug-resistant Klebsiella pneumoniae producing the extended-spectrum beta-lactamase CTX-M15 affected 247 mainly elderly patients in more than 30 wards in a 1000-bedded swedish teaching hospital between May 2005 and August 2007. A manual search of the hospital administrative records for possible contacts between cases in wards and outpatient settings revealed a complex chain of transmission. Faecal screening identified twice as many cases as cultures from clinical samples. Transmission occurred by direct and indirect patient-to-patient contact, facilitated by patient overcrowding. Interventions included formation of a steering group with economic power, increased bed numbers, better compliance with alcohol hand disinfection and hospital dress code, better hand hygiene for patients and improved cleaning. The cost of the interventions was estimated to be euro3 million. Special infection control policies were not necessary, but resources were needed to make existing policies possible to follow, and for educational efforts to improve compliance.

Building a concept system to structure the contents of a decision support system--a grounded theory study of concepts in the knowledge domain of urinary tract infection.

In this study we present a concept system for the knowledge domain of urinary tract infections. The system was constructed using grounded theory methods, sampling from Swedish reference texts and patient records from consultations for urinary tract infections. The concept system is intended for use as an aid in the construction of a decision support system (DSS) for urinary tract infections, and as a search module in the mentioned DSS. In total 561 concepts were categorized in four major, and 58 subordinate categories. Relationships between the more important categories were developed, and the frequency of use of the most common concepts is presented. Using the presented categorical structure as domain model a prototype DSS for dipslide urine cultures has been developed.

Performance of Uricult Trio assessed by a comparison method and external control panels in primary healthcare.

Using the comparison method, we have evaluated the technical performance of Uricult Trio by culturing on Uricult Trio and agar plates. Urine samples (477) from patients in primary healthcare were cultured in parallel in a microbiology laboratory. The result for Uricult Trio evaluated using the comparison method was incorrect in 32% of the cultures. We also studied the performance of Uricult Trio when used in primary healthcare by using external control panels. External control panels consisting of Uricult Trio, inoculated with known concentrations of certain bacterial strains, were used to assess the performance of Uricult Trio in primary healthcare during the period 1993-7. Aberrations in reports of concentration have ranged from 10% to 33%, failure in reporting of mixed culture from 0% to 91% and reporting of E. coli from 15% to 86%. There has been no sign of improvement over the years. The results indicate that Uricult Trio is unsuitable for indications other than exclusion of urinary tract infection or diagnosis of urinary tract infection caused by E. coli. Further, there is need for quality assurance and training activities at primary healthcare laboratories, probably best carried out in collaboration with local clinical microbiology laboratories.

Evaluation of two methods for improving quality of diagnosis of bacteriuria by culture in primary healthcare.

This study evaluates the effect of training on the results from Uricult Trio and an established urine culture when used at primary healthcare laboratories in two Swedish counties, Uppsala and Värmland. Urine cultures and dipslides, Uricult Trio, performed at these laboratories were interpreted a second time at central laboratories. Interpretation errors at the primary healthcare laboratories were calculated. Primary healthcare laboratories also received external control panels with urine cultures and dipslides. There was one study period each year for 3 years in Uppsala and for 2 years in Värmland. A training programme was completed between study periods in Värmland. In Uppsala, primary healthcare laboratory results could be reviewed, as interpretations by the central laboratory were returned to them. The main outcome measures were the percentage of interpretation errors which, in the first study period, was 33-39%. This dropped to 15-19% in the second study period. In the results from the external control panels there were no striking differences between the studied areas and Sweden as a whole, except that Uppsala showed a better result in reporting E. coli and failed in 10% compared to Sweden 46%. A method for both quality assessment and education is to ask the primary healthcare laboratories to send cultures to the central laboratory for interpretation requesting their return to the primary healthcare laboratory with the interpretation from the central laboratory attached.

Concepts, contexts and expert systems.

This paper describes problems identified in our attempts to develop an expert system for management of urinary tract infections. We found three aspects which we believe are important to consider when developing such systems. The objective of our future work will be to evaluate the impact of these problems on expert system development and usage.

Young women with symptoms of urinary tract infection. Prevalence and diagnosis of chlamydial infection and evaluation of rapid screening of bacteriuria.

OBJECTIVE: To estimate the prevalence of chlamydial infection among young women with UTI symptoms. To evaluate chlamydia diagnostics with the aid of enzyme immuno assay (EIA) on first-void urine. To evaluate rapid screening of bacteriuria, including low concentrations of common pathogens. DESIGN: EIA for detection of Chlamydia trachomatis antigen and confirmation with immunofluorescence test (DFA) in urine, cervical and urethral chlamydia culture, nitrite and granulocyte esterase test, urine sediment, chamber count, dipslide and conventional urine culture were performed. SETTING: Primary health care (PHC). PATIENTS: 217 women aged 15-35 years attending PHC for dysuria or urgency-frequency. MAIN OUTCOME MEASURES: Frequency of chlamydial infections. Sensitivity, specificity, predictive values of EIA and bacteriuria screening tests, respectively. RESULTS: The frequency of chlamydial infection was 3.7%. In spite of a high specificity of the EIA test (0.94 without DFA) the number of false positives exceeded the number of true positives. No single bacteriuria test showed sufficient diagnostic efficiency. CONCLUSIONS: Routine chlamydia testing in young women with UTI symptoms is recommended. EIA test on urine is of little use. Assessing diagnosis of UTI symptoms requires insight into the use of several rapid tests and a dialogue with the patient.

Meningitis due to Gemella haemolysans after neurosurgical treatment of trigeminal neuralgia.

We report a case of meningitis due to Gemella haemolysans. An 82-year-old man with trigeminal neuralgia and senile dementia developed signs of meningitis 24 h after retrogasserian glycerol injection. Cerebrospinal fluid cultures grew G. haemolysans. Initial treatment with ampicillin and cefotaxime followed by benzylpenicillin was successful.