Patient-tailored antibiotic therapy in acute cellulitis

Background Cellulitis is a bacterial skin infection characterised by pain, redness, swelling and fever. The infection frequently recurs and can lead to chronic tissue damage. It is one of the commonest infections leading to hospitalisation, accounting for ~90,000 admissions in England in 2018. The incidence of cellulitis in older populations is increasing as are the associated healthcare costs.

Although short treatment durations (5-7 days) are effective for most patients, some need longer to achieve sustained cure.

We are currently unable to identify who these patients are, which results in many patients receiving >14 days of treatment (probably more than twice the amount necessary).

This accelerates the development of antimicrobial resistance, which is a major threat to global health and is responsible for 33,000 deaths/year in Europe alone.

Aim To understand how patient factors, including early clinical response to treatment, can be used to tailor antibiotic treatment duration for cellulitis, giving the best chance of achieving cure while avoiding unnecessary antibiotic exposure.

Objectives: To develop a risk score to predict clinical outcome from baseline patient factors based on large-scale electronic health records (EHR).

To test this risk score in a prospective cohort of patients with cellulitis and investigate whether early clinical response adds prognostic value.

To identify the best technology for measuring skin temperature change as an objective marker of early clinical response. To develop an algorithm to stop antibiotics based on the risk score and clinical response.

Methods Retrospective EHR analysis: Data on ~5000 patients identified from an existing database based on diagnostic codes for cellulitis will be used to develop a risk score to predict clinical outcome from baseline factors. Prospective cohort study: Adults with cellulitis receiving antibiotic treatment from a hospital will be eligible.

The cohort will be used to test the risk score and investigate whether changes in affected skin temperature or other measures of early clinical response add prognostic value to the risk score. The primary outcome is sustained cure at 90 days.

Secondary outcomes include cure at 28 days, patient-reported improvement in symptoms, readmissions, mortality, antibiotic-related adverse events and quality of life.

Technology comparison: Three technologies will be used to measure skin temperature change (as an objective marker of early clinical response) in the cohort.

The technologies will be compared in terms of repeatability, reproducibility, time and cost to establish the optimal device for measuring clinical response in cellulitis.

Anticipated impact My goals are to develop a robust risk score to predict clinical outcome in cellulitis and an algorithm based on this risk score to tailor antibiotic treatment duration for patients with cellulitis.

The cohort will provide estimates of the rates of my outcomes to help design a trial to compare this approach to standard of care antibiotic duration. This will enable doctors to make evidence-based, personalised decisions about antibiotic treatment.

Reducing unnecessary antibiotic treatment will protect patients from antibiotic-associated adverse events and antibiotic-resistant infections.

For the wider healthcare system, this will make delivery of care more efficient and reduce costs for treating cellulitis.