The accuracy of fully automated algorithms for surveillance of healthcare-associated urinary tract infections in hospitalized patients

BACKGROUND

Surveillance for healthcare-associated infections such as healthcare-associated urinary tract infections (HA-UTI) is important for directing resources and evaluating interventions. However, traditional surveillance methods are resource-intensive and subject to bias.

AIM

To develop and validate a fully automated surveillance algorithm for HA-UTI using electronic health record (EHR) data.

METHODS

Five algorithms were developed using EHR data from 2979 admissions at Karolinska University Hospital from 2010 to 2011: (1) positive urine culture (UCx); (2) positive UCx + UTI codes (International Statistical Classification of Diseases and Related Health Problems, 10^th revision); (3) positive UCx + UTI-specific antibiotics; (4) positive UCx + fever and/or UTI symptoms; (5) algorithm 4 with negation for fever without UTI symptoms. Natural language processing (NLP) was used for processing free-text medical notes. The algorithms were validated in 1258 potential UTI episodes from January to March 2012 and results extrapolated to all UTI episodes within this period (N = 16,712). The reference standard for HA-UTIs was manual record review according to the European Centre for Disease Prevention and Control (and US Centers for Disease Control and Prevention) definitions by trained healthcare personnel.

FINDINGS

Of the 1258 UTI episodes, 163 fulfilled the ECDC HA-UTI definition and the algorithms classified 391, 150, 189, 194, and 153 UTI episodes, respectively, as HA-UTI. Algorithms 1, 2, and 3 had insufficient performances. Algorithm 4 achieved better performance and algorithm 5 performed best for surveillance purposes with sensitivity 0.667 (95% confidence interval: 0.594-0.733), specificity 0.997 (0.996-0.998), positive predictive value 0.719 (0.624-0.807) and negative predictive value 0.997 (0.996-0.997).

CONCLUSION

A fully automated surveillance algorithm based on NLP to find UTI symptoms in free-text had acceptable performance to detect HA-UTI compared to manual record review. Algorithms based on administrative and microbiology data only were not sufficient.

Risk factors for community-onset bloodstream infection with extended-spectrum β-lactamase-producing Enterobacteriaceae: national population-based case-control study

OBJECTIVES

The aim was to investigate risk factors for community-onset bloodstream infections with extended-spectrum β-lactamase-producing Enterobacteriaceae (EPE BSI).

METHODS

It is mandatory to report EPE BSI to a national register at the Public Health Agency of Sweden. Using this register, we performed a population-based case-control study from 2007 to 2012 of 945 cases and 9390 controls. Exposure data on comorbidity, hospitalization, in- and outpatient antibiotic consumption and socio-economic status were collected from hospital and health registers.

RESULTS

The overall incidence of EPE BSI was 1.7 per 100 000 person-years. The 30-day mortality was 11.3%. Urological disorders inferred the highest EPE BSI risk, adjusted odds ratio (aOR) 4.32 (95% Confidence Interval (CI) 3.41-5.47), followed by immunological disorders, aOR 3.54 (CI 2.01-6.23), haematological malignancy, aOR 2.77 (CI 1.57-4.87), solid tumours, aOR 2.28 (1.76-2.94) and diabetes, aOR 2.03 (1.58-2.61). Consumption of fluoroquinolones or mostly non-EPE-active antibiotics with selective Gram-negative spectrum of activity within the previous 3 months was associated with EPE BSI, aORs 5.52 (CI 2.8-11.0) and 3.8, CI 1.9-7.7) respectively. There was a dose-response relationship in EPE BSI risk with increasing number of consecutive regimens. Antibiotic consumption >3 months before EPE BSI was not associated with increased risk. Higher age, malignancies and education ≤12 years (aORs >2) were associated with increased 30-day mortality.

CONCLUSIONS

Targeted interventions should be directed towards improving care for patients with immunosuppression, urological disorders and subjects with lower socio-economic status. Antibiotic stewardship should focus on reduction of fluoroquinolones.

Invasive infection caused by Klebsiella pneumoniae is a disease affecting patients with high comorbidity and associated with high long-term mortality

Klebsiella pneumoniae (KP) is after Escherichia coli (EC) the most common gram-negative species causing invasive infections. Herein, we analyzed risk factors and prognosis in invasive infections caused by KP versus EC, in an area with low antimicrobial resistance. Moreover, we compared antimicrobial resistance and relative prevalence of KP and EC (KP/EC-ratio) in different European countries, using EARS-Net data. Adult patients admitted to Karolinska University Hospital 2006–2012 with invasive infection caused by KP (n = 599) were matched regarding sex and age with patients infected by EC. The medical records were retrospectively reviewed. Comorbidity was adjusted for with multivariable analysis. European data were retrieved from the EARS-Net database. No differences were observed in 7- and 30-day mortality between the groups. The 90-day mortality was significantly higher in the KP cohort (26% versus 17%, p<0.001), but not after adjusting for comorbidity. Malignancy was seen in 53% of the patients with KP versus 38% with EC, OR 1.86 (1.34–2.58). A significant increase in the rate of ESBL-production was observed in EC, but not in KP. The KP/EC-ratio remained stable. In contrast, European data showed increasing percentages of isolates non-susceptible to third-generation cephalosporins in EC and KP, and increasing KP/EC-ratio. Invasive infection caused by KP is a disease affecting patients with high comorbidity and associated with high 90-d mortality. The stable KP/EC-ratio and low occurrence of antimicrobial resistance in data from Karolinska University Hospital compared to aggregate data from 20 EARS-Net countries could be related to absence of clonal spread of multidrug-resistant KP.

Frequent acquisition of low-virulence strains of ESBL-producing Escherichia coli in travellers

OBJECTIVES

International travel is a risk factor for intestinal colonization with ESBL-producing Enterobacteriaceae (EPE). This prospective cohort study focuses on molecular features of and risk factors for travel-acquired EPE.

METHODS

Rectal swabs and survey data were collected from 188 Swedes travelling to four regions of high EPE prevalence. Samples were plated onto selective agars. ESBL producers were determined using phenotypic methods. Molecular characterization regarding virulence factors and phylogenetic grouping of ESBL-producing Escherichia coli was done using PCR. Isolates were also screened for the plasmid-mediated colistin resistance gene mcr-1.

RESULTS

Among 175 pre-travel EPE-negative participants, 32% were positive upon return. No carbapenemase-producing Enterobacteriaceae were found, but one CTX-M-producing E. coli harboured mcr-1 (travel to Thailand). Most E. coli strains (43.1%) belonged to phylogroup A and were rarely associated with extraintestinal infections and a few (9.2%) expressed uropathogenicity pap genes. During 10-26 months of follow-up, no clinical infections were observed. Colonization rates varied by visited region: the Indian subcontinent, 49.2%; northern Africa, 44.0%; South-East Asia, 19.1%; and Turkey, 9.5%. Travellers' diarrhoea (OR 2.5, P = 0.04) or antimicrobial treatment during the trip (OR 5.9, P = 0.02) were both independent risk factors for EPE colonization.

CONCLUSIONS

EPE acquired during travel have seemingly low pathogenicity, possibly indicating a low risk of clinical infection. Pre-travel advice should emphasize avoiding unnecessary antibiotic treatment during travel.