Making Blood Cultures Great Again: Quality Improvement Focus in the Emergency Department can improve the pre-analytical blood culture pathway
Optimising the blood culture pathway has the potential to improve antimicrobial stewardship (AMS) through greater opportunity to use targeted pathogen-directed antimicrobial therapy. An inter-professional blood culture improvement group was formed at Nottingham University Hospitals in August 2020, to improve the pre-analytical blood culture pathway in the Emergency Department (ED), which takes approximately 15,000 blood cultures per year. The group has representation from the Microbiology laboratory, Sepsis team, AMS team, ED staff and clinical skills educators.
Our aims are to:
Increase average blood culture fill volume to 8-10ml, to improve pathogen yield
Reduce contamination rates to ≤ 3%, to reduce waste in laboratory resources
We collect Quarterly performance data on fill volume, pathogen yield and contamination rate are regularly fed back to frontline staff and senior managers. Baseline data showed huge scope for improvement; the median fill volume in ED was only 3.4mL and contamination rates were 6.3%.
Audits of clinical practice identified correct technique and availability of equipment as targets for improvement. We produced videos of the correct technique and implemented changes to equipment, so that butterfly needles and cannula extension kits were available. A programme of training was delivered to ED staff, using interactive activities and series of engagement events. Training was delivered to all clinical staff in ED through clinical practice days and ‘shop floor’ teaching.
24-7 blood culture loading and processing was implemented in December 2020, reducing the time-to-pathogen identification and antibiotic susceptibility results. This means blood culture results are often available for patients still in ED, so that the benefits of improvements are seen by ED clinicians first-hand.
An external grant was awarded by the UK Sepsis Trust and Pfizer, to appoint a full-time Blood Culture Nursing Associate so that improvements can be consolidated and disseminated to other clinical areas.
Since the formation of the blood culture improvement group, ED blood culture fill volume have increased from 3.4mL to 6.5mL, which has led to an increase in pathogen yield from 7.5% to 10.1%. This equates to an additional 100 positive blood cultures per quarter; these patients benefit from targeted pathogen-directed antibiotic therapy (rather than broad-spectrum empirical antibiotics) and receive an infection specialist’s input into management so their care can be optimised e.g. switching to narrow-spectrum antibiotics, oral switch, advice on shortest effective course, targeted imaging and source control.
Due to blood culture laboratory 24/7 working, antibiotic susceptibility results are now available on average 12hrs earlier, allowing optimisation of antibiotics sooner in the patient journey. The majority of antibiotic susceptibility results are now available prior to morning clinical ward rounds, so that senior decision-makers can make antibiotic-related decisions sooner in the patients’ hospital stay.
Contamination rates fell from 6.3% to 4.4% equating to 75 fewer contaminants per quarter, reducing unnecessary laboratory work and freeing up scientific staff time to prioritise high value specimens. This also reduces clinical uncertainty and antibiotic use when skin contaminants are of uncertain significance.
Joint working on blood culture improvement has unified the Sepsis and Antimicrobial Stewardship agendas locally, reducing the risks of ‘mixed messaging’ around antibiotics by focussing on the highest risk group to improve antimicrobial stewardship in. As we continue to increase the number of patients with sepsis that have positive blood cultures, the gains in appropriate antibiotic use will exemplify that even the most acute patients with infection can receive narrow-spectrum antibiotics where microbiology testing supports this.
There are still further improvements to be made in ED to reach our target fill volume and contamination rates, which will be supported by a dedicated Sepsis nurse in ED working alongside the wider Blood Culture Improvement Group. Our next focus in ED is to improve the proportion of patients who have two sets of cultures taken prior to antibiotics to further maximise pathogen yield.
We have successfully procured software that will provide real-time blood culture performance data, replacing the manual audit methodology currently used, to support ward-level and specialty-level improvement work in clinical areas across the hospital.
Another area the group is working on is improvements in the time-to-loading of blood cultures, so that we can ensure blood cultures are loaded in the laboratory in <4hrs of being obtained wherever they are taken across our hospitals. This has the potential to further reduce the time-to-targeted antibiotic therapy. With the introduction of e-prescribing to our hospitals in late 2022, we are planning to use e-prescribing data to analyse in more detail the impact of blood culture positivity on time-to-targeted antibiotics and time-to-oral switch (which currently is not possible with paper prescribing). We recognise the importance of strengthening the governance around the pre-analytical blood culture pathway and we are currently working with the divisional Infection Prevention and Control leads to embed blood culture pathway performance into existing governance arrangements.
Evaluation of a sore throat test and treat service in community pharmacies: a tale of two approaches
In Wales, a national Sore Throat Test and Treat (STTT) service was commissioned in 2018 in 56 community pharmacies. The structured STTT pathway included screening of patients using FeverPAIN/CENTOR clinical assessment scores, a point-of-care test (POCT) for Group A Streptococcus (GAS) if threshold scores were met, and antibiotic supply in the pharmacy, in line with NICE guidance.
Evaluation of the pilot looking at data from a range of sources was positive and led to national roll-out being agreed by all health boards in Wales. A total of 134 pharmacies were offering the service by March 2020, when it was interrupted due to COVID-19. A new service delivery model was subsequently agreed, to safeguard patients and pharmacists. The change still allowed antibiotic supply to patients with threshold clinical scores, but removed the requirement for routine use of POCT to detect GAS.
This project is a post-pilot multifactorial evaluation and has focussed on two aspects: a) addressing the key evidence gaps from the pilot, specifically the longer-term impact of the service, once it was established in more geographical areas in Wales. A cross-sectional descriptive study was completed over 16 months, with the benefit of a much larger, more representative dataset for our analysis; b) addressing a research need as identified by NICE, who in their guidance highlighted that, although stratification of patients with sore throat symptoms based on clinical scoring is well-supported by evidence, the role of an additional GAS POCT is less clear, particularly in settings outside of GP practices, such as community pharmacy. The COVID delivery model provided us with the opportunity for extending the evaluation to consider this aspect.
To achieve our aims for both aspects of this project, individual-level STTT consultation data were extracted from electronic pharmacy records on the national IT platform, and analysed.
1) Over 16 months, 11,304 consultations were completed with a 21.3% antibiotic supply rate. All patients were provided with safety netting. This is the largest description of a pharmacy-led STTT service to date, and the overall rate of ~1/5 patients supplied with antibiotics is significantly lower to the figures reported as resulted from consultations with GPs. STTT continues to offer a safe option for patients; it can be delivered at scale to align with a pre-specified pathway that promotes appropriate use of POCT and antibiotics.
2) To look at the value of POCT, we compared data for the pre-pandemic period, between Nov18 and Sep19, and the COVID period, between Nov20 and May21. We found that the overall antibiotic supply rate increased by 27% when routine POCT was removed. Our analysis suggests that for every 100 STTT consultations with patients with the threshold clinical scores, the use of POCT may spare up to 36 courses of antibiotics. This increases to up to 47 courses spared when only those with higher clinical scores are included. This has significant implications for antimicrobial stewardship and is the rationale behind the re-introduction for the requirement for routine POCT (Dec21).
3) In our longitudinal study, 14.5% of all antibiotics supplied were for patients with FeverPAIN 2. For the same cohort of patients in the COVID delivery model, when pharmacists used their professional judgement to decide whether to offer POCT, they supplied less antibiotics than when using routine POCT. These findings, coupled with evidence on low rate of serious complications of upper respiratory tract infections in UK primary care, support possible further changes to the STTT delivery model. Suggestions include setting a threshold for re-consultation in the pharmacy (in lieu of delayed Abx) and/or changing criteria for POCT by increasing the threshold score from FeverPAIN>1 to >2.
STTT is rolling out to all areas in Wales currently. We are in the process of completing a matched cohort study involving data linkage, using the SAIL databank. This is the first time that routine pharmacy data is being linked to other sources such as routine GP and hospital data. It will enable us to follow patients on their journey through the NHS and study patient outcomes without using surrogate measures and making assumptions.