Association between antimicrobial usage in livestock and antimicrobial resistance in Escherichia coli isolates from human urinary tract infections in the Netherlands, 2009-2020.

BACKGROUND: In the last decade, veterinary antimicrobial usage (AMU) and antimicrobial resistance (AMR) among indicator bacteria in livestock have decreased substantially in the Netherlands. The extent to which this decrease has affected AMR levels among human infections remains unclear. OBJECTIVES: To assess the association between AMU in livestock and AMR in Escherichia coli isolates from human urinary tract infections (UTIs). METHODS: Data on AMR and AMU between 2009 and 2020 from Dutch national surveillance programmes for humans and livestock were used. Associations between AMU in four major livestock sectors and AMR in humans were assessed for 10 antimicrobial classes and the ESBL resistance profile, using logistic regression analysis. Associations between AMU and AMR in livestock, between AMR in livestock and in humans, and between AMU and AMR in humans were also assessed. RESULTS: Statistical significance was reached for 16/31 of the tested associations between AMU in livestock and AMR in human E. coli UTIs. Of the significant associations, 11 were positive (OR 1.01-1.24), whereas 5 were negative (OR 0.96-0.99). All associations between human AMU and AMR in E. coli isolates from UTIs were positive and statistically significant. Weak but significant positive correlations were also observed between livestock AMR and human AMR. CONCLUSIONS: Although several significant associations between AMU in livestock and AMR in human UTIs caused by E. coli were observed, the associations between AMU and AMR were generally stronger within the human and animal populations. This indicates that potential zoonotic spread of AMR in E. coli causing human UTIs from livestock sources is limited.

Faecal carriage of Clostridioides difficile is low among veterinary healthcare workers in the Netherlands.

Veterinary healthcare workers are in close contact with many different animals and might be at an increased risk of acquiring Clostridioides difficile. In this cross-sectional study, we assessed the prevalence and risk factors of C. difficile carriage in Dutch veterinary healthcare workers. Participants provided a faecal sample and filled out a questionnaire covering potential risk factors for C. difficile carriage. C. difficile culture positive isolates were polymerase chain reaction (PCR) ribotyped and the presence of toxin genes tcdA, tcdB and cdtA/cdtB was determined. Eleven of 482 [2.3%; 95% confidence interval (CI) 1.3-4.0] veterinary healthcare workers were carriers of C. difficile. Three persons carried C. difficile ribotype 078 (0.6%; 95% CI 0.2-1.8). Risk factors for carriage were health/medication and hygiene related, including poor hand hygiene after patient (animal) contact, and did not include occupational contact with certain animal species. In conclusion, the prevalence of C. difficile carriage in veterinary healthcare workers was low and no indications were found that working in veterinary care is a risk for C. difficile carriage.

Prolonged carriage of (livestock-associated) MRSA in individuals without professional livestock contact.

OBJECTIVES: To investigate prolonged carriage of MRSA in adults from the general population living in a livestock-dense area, using WGS. METHODS: A cross-sectional study during 2014-15 among 2492 adults without professional livestock contact identified 14 (0.6%) nasal MRSA carriers, 10 of which carried livestock-associated (LA)-MRSA of multiple-locus variable-number tandem repeat analysis (MLVA) complex (MC) 398. Two years later, 12 MRSA-positive and 88 MRSA-negative participants provided a second nasal swab and filled in a short questionnaire. Isolates from persons who were MRSA positive at both timepoints were compared using MLVA and isolates with the same MLVA type were sequenced. The WGS data were used for core-genome MLST (cgMLST) and resistome analysis, including sequenced isolates from the national MRSA surveillance. RESULTS: All MRSA-negative persons tested negative again, while 6 of the 12 initially MRSA-positive persons tested positive again. MLVA revealed that isolate pairs from five individuals had the same MLVA type, of which three were LA-MRSA. cgMLST showed that the distance between these isolate pairs ranged between 3 and 13 genes, while the minimum distance to unrelated isolates from the national MRSA surveillance was 38 genes. Moreover, the resistome present in the five isolate pairs was identical within each pair. None of the prolonged carriers was hospitalized during the 3 months before the sampling moment and none of them with LA-MRSA had contact with livestock in this period. CONCLUSIONS: Prolonged carriage of MRSA, including LA-MRSA, can be demonstrated after more than 30 months in persons without professional livestock contact.

Do vegetarians less frequently carry ESBL/pAmpC-producing Escherichia coli/Klebsiella pneumoniae compared with non-vegetarians?

BACKGROUND: ESBL and plasmid-mediated AmpC (pAmpC)-producing Enterobacteriaceae are frequently found on meat products in Dutch retail, especially on poultry. OBJECTIVES: We investigated whether vegetarians are at lower risk of carrying ESBL/pAmpC-producing Escherichia coli/Klebsiella pneumoniae (ESBL-E/K) compared with persons who consume meat. METHODS: Vegetarians, pescatarians (vegetarians who eat fish) and non-vegetarians (persons who eat meat at least three times per week) were asked to send in a faecal sample and a questionnaire. ESBL-E/K were cultured and MLSTs were determined. ESBL/pAmpC genes were analysed using PCR and sequencing. The risk of ESBL-E/K carriage in the three study groups was analysed using multivariable logistic regression. RESULTS: Prevalence of ESBL-E/K carriage was 8.0% in vegetarians (63/785; 95% CI 6.3-10.1), 6.9% in pescatarians (27/392; 95% CI 4.8-9.8) and 3.8% in non-vegetarians (14/365; 95% CI 2.3-6.3). Multivariable analysis showed an OR for ESBL-E/K carriage of 2.2 for vegetarians (95% CI 1.2-4.0) and 1.6 for pescatarians (95% CI 0.8-3.2) compared with non-vegetarians. The predominant MLST was E. coli ST131 and the most common ESBL genes were blaCTX-M-15, blaCTX-M-27, blaCTX-M-14 and blaCTX-M-1 in all diet groups. Independent risk factors for ESBL-E/K carriage were travel to Africa/Latin America/Asia (OR 4.6; 95% CI 2.8-7.7) in the past 6 months and rarely/never washing hands before food preparation (OR 2.5; 95% CI 1.2-5.0). CONCLUSIONS: Vegetarians and pescatarians did not have a lower risk of ESBL-E/K carriage compared with non-vegetarians, indicating that eating meat is not an important risk factor for ESBL-E/K carriage.

How does additional diagnostic testing influence the initial diagnosis in patients with cognitive complaints in a memory clinic setting?

BACKGROUND: patients suspected of dementia frequently undergo additional diagnostic testing (e.g. brain imaging or neuropsychological assessment) after standard clinical assessment at a memory clinic. This study investigates the use of additional testing in an academic outpatient memory clinic and how it influences the initial diagnosis. METHODS: the initial diagnosis after standard clinical assessment (history, laboratory tests, cognitive screening and physical and neurological examination) and the final diagnosis after additional testing of 752 memory clinic patients were collected. We specifically registered if, and what type of, additional testing was requested. RESULTS: additional testing was performed in 518 patients (69%), 67% of whom underwent magnetic resonance imaging, 45% had neuropsychological assessment, 14% had cerebrospinal fluid analysis and 49% had (combinations of) other tests. This led to a modification of the initial diagnosis in 17% of the patients. The frequency of change was highest in patients with an initial non-Alzheimer's disease (AD) dementia diagnosis (54%, compared with 11 and 14% in patients with AD and 'no dementia'; P < 0.01). Finally, after additional testing 44% was diagnosed with AD, 9% with non-AD dementia and 47% with 'no dementia'. CONCLUSION: additional testing should especially be considered in non-AD patients. In the large group of patients with an initial AD or 'no dementia' diagnosis, additional tests have little diagnostic impact and may perhaps be used with more restraint.

Temporal association of antimicrobial use in livestock with antimicrobial resistance in non-typhoid Salmonella human infections in the Netherlands, 2008-2019.

Background: Antimicrobial use (AMU) in livestock contributes to antimicrobial resistance (AMR) among zoonotic pathogens, such as non-typhoid Salmonella (NTS). Since 2009, the Netherlands has made substantial efforts to reduce AMU in livestock. Objectives: To assess the association between AMU in livestock and AMR in NTS human isolates. Additionally, associations between AMU in broilers/pigs and AMR in NTS broiler/pig isolates, and between AMR in broilers/pigs and in human NTS isolates were assessed. The focus was on Salmonella Enteritidis (SE) and Salmonella Typhimurium including its monophasic variant (ST/STM). Methods: A national population registry-based study was conducted in the Netherlands from 2008 to 2019. Multivariable logistic regression models were used to assess the associations between livestock AMU and NTS resistance proportion in humans and broilers/pigs, overall as well as per class-specific antimicrobials. Correlation analysis was performed to relate AMR proportions between human and broiler/pig NTS isolates. Results: For SE, only a positive association between penicillins use in broilers and resistance to ampicillin among human isolates was significant. For ST/STM, most associations between AMU in livestock and AMR among human isolates were significantly positive, overall and per class-specific antimicrobials, namely for penicillins-ampicillin, tetracyclines-tetracycline and sulfonamides/trimethoprim-sulfamethoxazole/trimethoprim. Significantly positive associations between AMU in broilers/pigs and AMR in broiler/pig ST/STM isolates were also observed, but not between broiler/pig and human AMR levels. Conclusions: Significant associations were generally found between livestock AMU and AMR in human and broiler/pig ST/STM isolates. However, confounding factors, such as imported meat and travel are of concern. To fully comprehend the impact of livestock AMU on resistance in human NTS isolates, it is imperative to enhance AMR surveillance of NTS.

ESBL/pAmpC-producing Escherichia coli and Klebsiella pneumoniae carriage among veterinary healthcare workers in the Netherlands.

BACKGROUND: Animals are a reservoir for ESBL/pAmpC-producing Escherichia coli/Klebsiella pneumoniae (ESBL-E/K). We investigated the association between occupational contact with different types of animals and the prevalence of ESBL-E/K carriage among veterinary healthcare workers, assessed molecular characteristics of ESBL-E/K, and followed-up on the ESBL-E/K carriage status of participants and their household members. METHODS: Participants completed a questionnaire about their contact with animals at work and at home, health status, travel behaviour and hygiene, and sent in a faecal sample which was tested for the presence of ESBL-E/K. Resistance genes were typed using PCR and sequencing. ESBL-E/K positive participants and their household members were followed up after 6 months. Risk factors were analysed using multivariable logistic regression methods. RESULTS: The prevalence of ESBL-E/K carriage was 9.8% (47/482; 95%CI 7.4-12.7). The most frequently occurring ESBL genes were blaCTX-M-15, blaCTX-M-14 and blaDHA-1. The predominant sequence type was ST131. None of the occupation related factors, such as contact with specific animal species, were significantly associated with ESBL-E/K carriage, whereas travel to Africa, Asia or Latin America in the past 6 months (OR 4.4), and stomach/bowel complaints in the past 4 weeks (OR 2.2) were. Sixteen of 33 initially ESBL-E/K positive participants (48.5%) tested positive again 6 months later, in 14 persons the same ESBL gene and E. coli ST was found. Four of 23 (17.4%) household members carried ESBL-E/K, in three persons this was the same ESBL gene and E. coli ST as in the veterinary healthcare worker. CONCLUSIONS: Despite the absence of specific occupation related risk factors, ESBL-E/K carriage in veterinary healthcare workers was high compared to the prevalence in the general Dutch population (5%). This indicates that occupational contact with animals is a potential source of ESBL-E/K for the population at large.

The effect of body mass index on the risk of surgical site infection.

OBJECTIVE: Obesity is considered a risk factor for surgical site infection (SSI). We quantified impact of body mass index (BMI) on the risk of SSI for a variety of surgical procedures. METHODS: We included 2012-2017 data from the Dutch national surveillance network PREZIES on a selection of frequently performed surgical procedures across different specialties. Patients were stratified into 5 categories: underweight (BMI, <18.5 kg/m2), normal weight (BMI, 18.5-25), overweight (BMI, 25-30), obese (BMI, 30-40) and morbidly obese (BMI, ≥40). Multilevel log binomial regression analyses were performed to assess the effect of BMI category on the risk of superficial, deep (including organ-space) and total SSI. RESULTS: Of the 387,919 included patients (ranging from 2,616 for laparoscopic appendectomy to 119,834 for total hip prosthesis), 3,676 (1%) were underweight, 116,778 (30%) had normal weight, 154,339 (40%) were overweight, 104,288 (27%) had obesity, and 8,838 (2%) were morbidly obese. A trend of increasing risk of SSI when BMI increased from normal to morbidly obese was observed for almost all surgery types. The increase was most profound in surgeries with clean wounds, with relative risks for morbidly obese patients ranging up to 7.8 (95% CI, 6.0-10.2) for deep SSI in total hip prosthesis. In chest and abdominal surgeries, the impact was larger for superficial SSI than for deep SSI. CONCLUSIONS: The results of our research provide evidence for the need of preventive programs targeting SSI in overweight and obese patients, as well as for the prevention of obesity in the general population.

Contribution of Prior, Multiple-, and Repetitive Surgeries to the Risk of Surgical Site Infections in the Netherlands.

OBJECTIVE Surveillance is an important strategy to reduce the incidence of surgical site infections (SSIs). We investigated whether prior, multiple-, or repetitive surgeries are risk factors for SSI and whether they should be preserved in the protocol of the Dutch national SSI surveillance network. METHODS Dutch national SSI surveillance data 2012-2015 were selected, including 34 commonly performed procedures from 8 major surgical specialties. Definitions of SSIs followed international standardized criteria. We used multivariable multilevel logistic regression techniques to evaluate whether prior, multiple-, or repetitive procedure(s) are risk factors for SSIs. We considered surgeries clustered within partnerships of medical specialists and within hospitals (random effects) and different baseline risks between surgical specialties (fixed effects). Several patient and surgical characteristics were considered possible confounders and were included where necessary. We performed analyses for superficial and deep SSIs combined as well as separately. RESULTS In total, 115,943 surgeries were reported by 85 hospitals; among them, 2,960 (2.6%) resulted in SSIs (49.3% deep SSIs). The odds ratio (OR) for having prior surgery was 0.94 (95% confidence interval [CI], 0.74-1.20); the OR for repetitive surgery was 2.39 (95% CI, 2.06-2.77); and the OR for multiple surgeries was1.27 (95% CI, 1.07-1.51). The latter effect was mainly caused by prolonged duration of surgery. CONCLUSIONS Multiple- and repetitive surgeries significantly increased the risk of an SSI, whereas prior surgery did not. Therefore, prior surgery is not an essential data item to include in the national SSI surveillance network. The increased risk of SSIs for multiple surgeries was mainly caused by prolonged duration of surgery, therefore, it may be sufficient to report only duration of surgery to the surveillance network, instead of both (the variables duration of surgery and multiple surgeries). Infect Control Hosp Epidemiol 2017;38:1298-1305.