- Volume 2, Issue 2, 2020

Dengue is caused an arbovirus and transmitted by Aedes mosquito. The mosquito lifecycle is influenced by various climatic factors This study was carried out to examine whether the climatic factors data can be used to predict yearly dengue cases .

Monthly reported dengue cases and climate data for the years 2012–2016 were obtained from the Chief Medical and Health officer,Bhopal and Meteorological Department , respectively. One-way analysis of variance was used to analyse whether the climatic parameters differed significantly among seasons. Four models were developed using negative binomial generalized linear model analysis. Monthly rainfall, temperature, were used as independent variables, and the number of dengue cases reported monthly was used as the dependent variable. The first model consider data from the same month, while the other three models involved incorporating data with a lag phase of 1, 2, and 3 months, respectively.

Climatic factors, rainfall and maximum temperature were significantly correlated with monthly dengue cases. The greatest number of cases was reported during the post-monsoon period. Temperature, rainfall, and humidity varied significantly across the pre-monsoon, monsoon, and post-monsoon periods. The best correlation between these climatic factors and dengue occurrence was at a time lag of 2 months.

Conclusions:

The climate had a major effect on the occurrence of dengue infection in Bhopal city of central India. Though the prediction model had some limitations in predicting dengue cases, it could forecast possible outbreak two months in advance with considerable accuracy, and can act as early warning system.

Antimicrobial stewardship ward rounds and phone advice are fundamental to improvement of infection treatment and prevention in hospitals. In response to a local and national shortage of consultant medical microbiologists, a pharmacist-led antimicrobial stewardship service was established.

Antimicrobial stewardship interventions in a large regional hospital were analysed from 8 January 2018 to 14 March 2019. Collaborative ward rounds were conducted with infection prevention and control nurses, and microbiology staff when available. Ad hoc ward rounds and phone interventions (via a dedicated “antibiotic advice hotline”) were also conducted, with most referrals coming from junior doctors and ward pharmacists. The commonest clinical areas visited were elderly care, respiratory and general surgery.

1507 antimicrobials were reviewed from 1006 interventions (16 per week). Antimicrobials most reviewed were piperacillin/tazobactam (n=152; 10%), metronidazole (n=152), co-amoxiclav (n=140; 9%), teicoplanin (n=137, 9%) and gentamicin (n=116; 8%). The commonest organisms were Escherichia coli (n=115), Staphylococcus aureus (n=95) and Clostridium difficile (n=52). The most common recommendations were intravenous to oral switch (n=185; 18%), continue (n=175; 17%), escalate (n=136; 14%) and review dose (n=97; 10%). Antimicrobials were optimised for discharge in 270 cases, through oral switch, home intravenous antimicrobial referral or cessation.

The results demonstrate the value of novel antimicrobial stewardship approaches that are required in today’s NHS given shifts in staff availability and recognising advanced clinical practice among non-medical staff. Future planned interventions will focus on improving the home intravenous antimicrobial referral and review process.

Assurance and intervention is needed around antimicrobial prescribing in secondary care outpatient settings, which has increased by 21% in 5-years. Use of NHS FP10 prescriptions impedes timely monitoring of prescribing. This project aimed to develop an efficient method for monitoring and reporting outpatient prescribing on FP10 forms.

Methods

A Microsoft Excel spreadsheet was developed to allow direct importing of antimicrobial prescribing data from ePACT2, which then automatically updates charts of prescribing activity using VLookup formulae and pivot charts. Reports were developed in Microsoft Word that directly pull updated charts from the spreadsheet described.

The method allows rapid calculation and collation of DDDs and creates charts for assessing monthly antimicrobial prescribing trends. The reports highlight areas for review and action, which have driven engagement in antimicrobial stewardship across the organisation. Unexpectedly, these reports heightened nurses’ awareness of antimicrobial stewardship (AMS), who now seek reassurance on, and challenge, antimicrobial prescribing at the time of prescribing. The process takes less than 20-minutes on average (versus 3-hours previously), freeing up antimicrobial pharmacists’ time to interrogate prescribing trends and undertake clinical duties. This method also allows potentially inappropriate prescriptions to be identified and pulled from NHS BSA sooner, so that individual prescribers and teams can be targeted for intervention and reflection.

This method allows efficient reporting of antimicrobial prescribing, enables stewardship teams to focus interventions on prescribers and teams, and drives engagement in antimicrobial stewardship. This method could be adapted by other organisations to monitor and interrogate antimicrobial prescribing.

Increasing antimicrobial resistance has renewed interest in older, infrequently used antimicrobials. Cotrimoxazole shows future promise; however, acute kidney injury (AKI) and hyperkalaemia are potential complications. Recognising risk factors for cotrimoxazole-associated AKI and hyperkalaemia, and quantifying the impact, are required for safe use against future antimicrobial resistance.

A single-centre retrospective observational study using electronic-healthcare records of patients prescribed cotrimoxazole was conducted. Patient risk factors were identified, and serum creatinine and potassium levels were analysed over the subsequent 21-days from prescription. Univariate and multiple logistic regression analyses were performed. The project was registered locally with the clinical governance team as service evaluation [Ref: CSS033].

Of 214 patients, 42 (20%) developed AKI and 33 (15.4%) developed hyperkalaemia. Low baseline eGFR (<60mls/min/1,73m2, OR = 7.78, 95%CI 3.57-16.13, p<0.0001) and pre-existing cardiac disorders (OR = 2.40, 95%CI 1.17-4.82, p=0.011) significantly predicted AKI. Early serum creatinine increases within 2-4 days of therapy predicted future AKI (OR = 3.65, 95%CI 1.73-7.41, p = 0.001). A low baseline eGFR also significantly predicted future hyperkalaemia (<60mls/min/1.73m2, OR = 6.80, 95%CI 3.09-15.06, p<0.0001). Low-dose cotrimoxazole (<1920mg/day) was associated with lower AKI and hyperkalaemia risk (p = 0.007 and 0.019, respectively).

Cotrimoxazole-associated AKI and hyperkalaemia is frequent and dose-dependant. Renal function and pre-existing cardiac disorders should be carefully evaluated before prescribing cotrimoxazole. Serum creatinine should be monitored in the first 2-4 days of treatment to identify susceptible patients, and low-doses considered if AKI or hyperkalaemia is suspected.

A blood stream infection (BSI) presents a complex and serious health problem and is growing in light of the eminent antimicrobial resistance threat. Progress must be made towards rapid BSI diagnosis and antimicrobial susceptibility testing (AST) to reduce preventable death in BSI.

Positive blood cultures from Ninewells Hospital, Dundee were studied prospectively. Flagged positive blood cultures were processed by Gram-staining, Vitek identification and AST or disc diffusion for ceftriaxone susceptibility and by scattered light integrated collection device (SLIC). Susceptibility to a panel of five antibiotics as defined by EUCAST breakpointswere compared.The time to AST result and AST categorical agreement was compared for standard methods and SLIC, any discrepancies were resolved by the EUCAST broth micro-dilution reference method.

A total of 505 bacterial-antimicrobial combinations were analysed. A categorical agreement of 95.45% (482/505) was achieved between SLIC and Vitek/disk diffusion. The remaining 23 discrepancies were resolved by broth micro-dilution with 10 AST results agreeing with the SLIC result and 13 in agreement of Vitek/disk diffusion. Thus the overall AST agreement of the standard workflow was 98.01% (495/505) compared to 97.43% (492/505) using SLIC. The mean time for AST was 1.94 + 0.02 h and 10.53 + 0.46 h for SLIC and Vitek respectively. Overall SLIC was calculated to save 25hrs from positive blood culture to AST outcome.

SLIC has the capacity to provide AST at the same time as Gram stain or MALDI-TOF identification.This could improve the value of blood culture in clinical practice.

We discuss the phenotypic features of a clinically significant blood culture isolate that was ultimately identified as Pasteurella bettyae, with important laboratory learning points.

The blood culture from a female patient with septic shock from an unknown source grew a Gram-negative, oxidase-positive, coccobacillus that was initially identified by API® NH as Haemophilus influenzae (%ID=99.8; T=0.75). A respiratory source was therefore investigated for. The MALDI-TOF result, which was returned to the laboratory 4 weeks after blood culture became positive, revealed the isolate to be Pasteurella bettyae (score=2.305). P. bettyae is an unusual organism associated with gynaecological infections, such as Bartholin’s abscess. Local source control, such as incision and drainage of a Bartholin’s abscess, may accelerate resolution of sepsis. The laboratory misclassification of P. bettyae as H. influenzae using API® NH may lead to misidentification of the source of infection.

1. Correctly identifying P. bettyae as the cause of a Gram-negative coccobacillus bacteremia can prompt suspicion of a gynaecological source of sepsis.

2. P. bettyae and H. influenzae, as members of the same family Pasteurellaceae, have overlapping morphological and biochemical features that can lead to laboratory misclassification. A catalase test and x+v plates can clearly differentiate the isolates (discussed further in poster). It is recommended that laboratories perform both tests, with MALDI-TOF confirmation if possible, rather than rely solely on the result of an API® NH.

Being labelled as “allergic” to an antimicrobial (commonly penicillins) can lead to worse clinical outcomes and increase antimicrobial resistance. Unclear or inappropriate allergy documentation contributes to this issue so clarifying “allergy” status is important to optimise clinical outcomes.

Methods

The study was conducted at St Thomas’ Hospital, a large teaching hospital in London. An antimicrobial allergy assessment process, consisting of a researcher-developed and administered questionnaire, a review of patient medical records and categorisation of the drug allergy ‘label’, was applied to eligible patients from selected wards. Recommendations for allergy ‘de-labelling’ and referrals were given based on patient categorisation.

Sixty assessments were completed during the study. Six patients (10%) were identified as unlikely to have a true allergy and could potentially be ‘de-labelled’. A further thirty-seven patients (61.7%) were identified as eligible for allergy referral and testing. The allergy ‘labels’ of twenty-seven patients (45%) led to them receiving second-line antibiotic therapy. Twenty-two patients (34.7%) had inconsistencies in their allergy documentation across the 3 main electronic clinical systems.

This study demonstrated that the use of a simple, standardized, pharmacy-led approach to antimicrobial allergy assessment could lead to some patients being 'de-labelled' outside of specialist allergy settings. The implementation of this approach could immediately bring about improvements in antimicrobial usage and patient outcome.

61.5% of patients had a CXR at diagnosis. 72.5% of patients had a CXR or a CT scan. 8 patients had an IGRA test performed, of which 2 were inconclusive. 20% of patients from Africa and 0% of patients from Asia had an IGRA test. Of those patients with a CD4 count below 200 cells/mm3 9.1% had an IGRA test.

Discussion

An assessment of patient TB risk may need to be undertaken to allow thorough latent TB screening using IGRA tests. Automatic IGRA tests could be done on all patients with a CD4 count <200 cells/mm3.

If the infectious focus is not identified, there is an associated higher mortality in Staphylococcus aureus bacteraemia (SAB) [1]. A retrospective observational study found patients with high-risk SAB who had a PET scan had a 67% reduction in mortality compared to those who did not [2]. We used electronic data to determine the range of infectious foci in SAB and the use of PET in our trust.

From 1/1/13 to 31/12/18 all patients with SAB at St Thomas’ Hospital, London, were reviewed by the infectious diseases team and data collected prospectively using an Access database. A retrospective analysis of this database, electronic records, radiology and nuclear imaging was conducted.

355 episodes of SAB affected 296 patients. 28 (8%) episodes were MRSA in 24 (8%) of patients. Infectious sources found included bone and joint (23%), IV access (15%), vascular infection (15%), no infective focus (13%), SSTI (12%) and endocarditis (9%). Imaging used to determine the focus of infection included MRI (18%), CT (26%) and PET scan (10%). In-hospital mortality was significantly lower in those who had a PET scan than in those who did not (1/31, 3% v. 50/265 patients (last episode), 19%, P=0.03).

Conclusions:

Mortality was lower in those who had PET at our trust. This is consistent with data published from other centres and requires further investigation. A prospective trial of PET in SAB is urgently needed.

References:

[1] PMID: 30179645

[2] PMID: 28336786

Paediatric antimicrobial stewardship (PAS) networks exist in the USA and Australia but not in the UK. We sought to capture a snapshot of the current PAS landscape across UK children’s hospitals.

A survey of PAS activities was conducted in tertiary children’s hospitals.

Infectious disease or microbiology consultants responded to the survey (n=15). All hospitals had neonatal, paediatric intensive care and surgical patients.All centres provided a PAS education programme for doctors, 7 for nurses and 9 for pharmacists as well. All centres had empirical antimicrobial prescribing guidelines. All centres with a paediatric infectious diseases (PID) team (11/15) used “audit and feedback” rounds, although their frequency and coverage varied. The PAS teams mostly included a PID consultant and/or microbiology consultants and a pharmacist. Three centres also had a nurse covering also the paediatric out-patient antibiotic treatment programme. Funding for PAS were inconsistent: Great Ormond Street Hospital had secured a dedicated full-time paediatric microbiologist, antimicrobial pharmacist and PID consultant with a ratio of 1/453 in-patient beds. 9 centres did not have dedicated funding for a paediatric antimicrobial pharmacist, 7 did not have funding for a paediatric infectious disease consultant. Only 2 hospitals had microbiology consultant time for paediatric audit and feedback.

PAS programmes in the UK are limited, funding is inconsistent and their set up is variable, even in tertiary children’s hospitals with a strong interest in infectious diseases. We propose a national PAS network to advocate for more consistency and research into the implementation of PAS programmes.

Fifty five percent of the Candida isolates produced biofilm. Biofilm producing abilities of clinically significant isolates (80%) was found to be significantly higher than commensals (20%). Biofilm positive Candida isolates were most commonly obtained from blood (34.5%). Biofilm production in non-albicans Candida (67.9%) was found to be significantly higher than C. albicans (38.6%). Majority of the biofilm positive Candida isolates produced Grade 2 (moderate) biofilm. Candida tropicalis accounted for maximum biofilm production comprising 20% Grade 4, 53.8% Grade 3 and 50% Grade 2 biofilm. Concordance in grading between the two methods was observed in 72.7% of the isolates. Spectrophotometric method was found to be more sensitive than visual method for detection of Candida biofilm.

The importance of studying Candida biofilms is to ascertain new therapeutics and techniques to manage these infections clinically and improve the outcome as these are associated with high morbidity, mortality and resistance to antifungal drugs.

Streptococcus pyogenes (Group A β -hemolytic Streptococcus) infections in humans range from mild pharyngitis to life-threatening infections such as septicaemia and necrotizing fasciitis. We assessed the diagnostic accuracy of the Xpert® Xpress Strep A Assay, a new automated real-time polymerase chain reaction (PCR) that can detect this pathogen.

Methods

We evaluated the diagnostic accuracy (sensitivity, specificity PPV and NPV) of Xpert® Xpress Strep A Test (Cepheid, Maurens-Scopont, France) for the detection of Group A Streptococcus in throat swabs using Copan virus transport medium at Ninewells Hospital & Medical School (Dundee, Scotland, UK). The performance of Xpert t® Xpress Strep A Assay was compared with routine culture (gold standard method).

A total of 323 clinical throat swab samples were tested. The sensitivity of Xpert ® Xpress Strep A Assay was found to be 95% (95% CI, 75.13% -99.87); specificity 96.5% (95% CI 93.83% -98.36%);PPV 65.5% (95% CI 50.61% -77.89%); and NPV 99%(95% CI 97.68% -99.95%). Overall, there were eleven discrepant results, of these 10 were positive by PCR and not confirmed by culture and one sample was recorded positive by culture and negative by PCR. The turnaround time (TAT) was 24 minutes for negative samples and 18 minutes for positive samples.

The Xpert® Xpress Strep A Assay for detection of Streptococcus pyogenes in throat swabs had high sensitivity, specificity and rapid TAT. Use of this technology as a point-of-care test could be explored further to assess its impact on antimicrobial prescribing and infection prevention and control measures.