New tests could cut unnecessary antibiotics

When a seriously unwell patient arrives at hospital with a respiratory illness, doctors often have to make treatment decisions quickly.

The infection might be viral, meaning antibiotics will not help. But with the possibility of a bacterial infection or sepsis, clinicians may prescribe them as a precaution.

Southampton researchers are investigating whether new rapid diagnostic tests could change this. The results could give doctors the confidence to avoid prescribing antibiotics that are not needed.

Preventing antibiotic resistant infections

Prescribing unnecessary antibiotics contributes to the development of resistant bacteria. These cause infections that cannot be treated with certain antibiotics, making them much harder to treat.

Professor Tristan Clark, Professor of Infectious Diseases, and Dr Nathan Brendish, Senior Clinical Lecturer, are part of a wider team evaluating new tests. They are both part of the NIHR Biomedical Research Centre (BRC): Southampton.

“If we can safely reduce unnecessary antibiotic use, we have a chance of reducing the development of resistance,” says Professor Clark.

The research brings together clinical and non-clinical staff, PhD students and colleagues across Southampton’s Faculty of Medicine and BRC.

A difficult decision for clinicians

Antibiotics treat bacterial infections, but do not work against viruses. Yet distinguishing between the two when a patient first arrives can be difficult.

“Even if a doctor thinks somebody probably has a viral infection and does not need antibiotics, they will often prescribe them just in case,” Professor Clark explains.

“We need tests that are fast enough to influence that decision, and accurate enough to reassure clinicians that it is safe to withhold antibiotics.”

The aim is not simply to prescribe fewer antibiotics. It is to give clinicians better evidence. This would enable them to provide the right treatment to the right patient at the right time.

Looking at the patient’s immune response

Established diagnostic tests work by detecting a specific cause of a respiratory disease. For example, they might detect a specific virus that causes flu or respiratory syncytial virus (RSV).

Yet detecting a virus does not necessarily rule out a simultaneous bacterial infection. It is also impractical to test for every possible respiratory virus.

Instead, the technologies being evaluated in Southampton examine specific markers in a patient’s blood. These are used to determine whether their immune response is more consistent with a viral or bacterial infection.

“You cannot easily test for every single virus,” says Dr Brendish. “These newer tests look at what the patient’s body is responding to instead.

“If a test can show that the immune system is fighting a viral infection rather than a bacterial one, that could give a clinician the confidence not to prescribe antibiotics.”

Equally, a result showing a bacterial infection would enable the patient to receive antibiotics early on.

Finding the best-performing test

The team is now completing a diagnostic accuracy study. This involves 500 patients with acute respiratory illness.

The study is comparing several host-immune response-based technologies. The aim is to establish test which can most accurately distinguish between viral and bacterial infections.

Most of these tests use blood taken from a vein. However, researchers have also evaluated a finger-prick test that can produce a result in around 10 minutes.

“We are evaluating a variety of tests to see which should be taken forward,” Dr Brendish says. “The decision has to come down to which test provides the best diagnostic accuracy.”

Taking rapid testing into NHS care

From winter next year, the researchers plan to evaluate the clinical impact of the best-performing test.

They intend to do this in a randomised controlled trial. This will involve adults who come to hospital with an acute respiratory illness.

The research will take place in areas such as the emergency department, where urgent prescribing decisions are made.

“We are looking at patients where there is just enough uncertainty for somebody to say, ‘You probably have a viral infection, but I’m going to give you antibiotics just in case’,” says Dr Brendish. “That is the unnecessary use we want to eliminate.”

The trial is expected to run over two winter seasons. It will examine whether rapid testing changes antibiotic prescribing.

The researchers will assess whether the test is safe. They will also determine if it can be incorporated effectively into clinical care.

The work forms part of Professor Clark's NIHR Research Professorship. It builds on previous Southampton studies. These explored how rapid diagnostics can improve treatment decisions.

For Professor Clark, the prospect of bringing the research back to the bedside is particularly exciting.

“In our previous trials, we have seen the difference that rapid diagnostic results can make to patients,” he says. “For example, during the COVID-19 pandemic, we showed how point-of-care tests improved patient care and reduced hospital acquired infection.”

“I’m really looking forward to finding out in the future trial whether we can make that difference for patients.”