New blood test 'could help prevent antibiotic misuse'

"A new blood test can help doctors tease out whether an infection is caused by a bacteria or a virus within two hours," BBC News reports. The test, which looks at protein pathways in the blood, could help to appropriately target the use of both antibiotics and antivirals.

In many cases, it is unclear whether a person’s symptoms are being caused by a viral or bacterial infection, and current testing can take up to several days to find out.

In cases of severe illness, antibiotics are usually prescribed while waiting for the results, and this can contribute to antibiotic resistance.

Israeli-based researchers who developed the test used 1,002 children and adults who had been admitted to hospital. The test was good at distinguishing between viral and bacterial infections, and separating people with and without an infectious disease.

However, it needs to be used by a greater number people, to test its effectiveness, and has not yet been used to influence treatment. Further research, including randomised controlled trials, will be required before it could be used in a clinical setting.

Where did the story come from?

The study was carried out by researchers from several institutes and medical centres in Israel. It was funded by MeMed, a company based in Israel that designs and manufactures diagnostic tests. Most of the researchers were employed by MeMed and some reported owning stock options with the company.

The study was published in the peer-reviewed medical journal PLOS One. It is published on an open-access basis, so is free to read online.

The research was accurately reported by BBC News.

What kind of research was this?

This was a laboratory study, which used blood samples from a cohort of patients admitted to hospital. It aimed to develop a blood test that could distinguish between viral and bacterial infections.

Overuse or incorrect use of antibiotics leads to the inadvertent selection of bacteria that have resistance to them. Over time, the resistant bacteria can become more common, making the drugs less useful.

This is causing global concern, as infections that have been easy to treat with antibiotics may now emerge as serious, life-threatening conditions. This can happen by people being given "broad spectrum antibiotics". This happens when an infection is suspected, but before any microbiological results can show the exact type of infection. This means that some people will be given the wrong antibiotic, too many antibiotics or an antibiotic for illness caused by viruses, which will be ineffective.

Current tests that can be speedily obtained when an infection is suspected include non-specific markers of infection and the number of different white blood cells. These cells are specialised to fight different types of infections, with neutrophils mainly fighting bacteria and lymphocytes mainly fighting viruses. However, the interpretation of these tests is not straightforward, as both can be increased in each type of infection.

The researchers wanted to develop a test that could show whether the infection is from a bacteria or virus, so that fewer

What did the research involve?

The researchers took blood samples from 30 people and measured a number of proteins that are produced by the immune system in response to bacterial or viral infections. They used this information to create a blood test that measured these proteins. They then tested how accurate it was in 1,002 children and adults admitted to hospital with or without a suspected infection.

They used a systematic literature review to identify 600 proteins that can increase during bacterial and viral infections. Using samples from 20 to 30 people, half of whom had a viral infection and half a bacterial infection, they whittled down the number of proteins that are distinctly raised in each type of infection to 86. They then looked at the level of these proteins in 100 people, half with each infection, and found that 17 of the proteins were the most useful. Using statistical programmes, they chose three proteins for their final test. These were:

  • CRP (C-Reactive Protein) – a protein that increases in response to tissue injury, infection and inflammation; this is routinely used in clinical practice
  • IP-10 (Interferon gamma-induced protein-10)
  • TRAIL (Tumour necrosis factor [TNF] related apoptosis-inducing ligand)

The researchers then used the test on blood samples from children and adults from two medical centres who were suspected of having an infection due to a fever of over 37.5C developing within 12 days of the onset of symptoms. A control group consisted of people who were not suspected of having an infection – such as people with suspected trauma, stroke or heart attack – or healthy people.

People were excluded who had:

  • evidence of acute infection in the previous two weeks
  • immune deficiency from birth
  • immunosuppressant treatment
  • cancer
  • HIV
  • hepatitis B or C

After all usual test results were obtained, a panel of three clinicians individually reviewed the clinical notes and test results, and recorded whether each person had a bacterial infection, viral infection, no infection, or that it was unclear. The three doctors made their assessment independently and were not told what the other doctors had decided, and did not know the result of the test in development. They compared findings from this expert panel with the results of their blood test.

What were the basic results?

A total of 765 participants were diagnosed with either a viral infection, bacterial infection or no infection. Additionally, there were 98 people who did not have a clear diagnosis.

The test was good at distinguishing between viral and bacterial infections, and separating people with and without an infectious disease. The test remained robust regardless of where the infection was, such as in the lungs or the gut, or variables such as age.

Results were not clearly presented for the 98 people without a firm clinical diagnosis.

How did the researchers interpret the results?

The researchers concluded that "the accurate differential diagnosis provided by this novel combination of viral- and bacterial-induced proteins has the potential to improve management of patients with acute infections and reduce antibiotic misuse".


This new test shows promising results in distinguishing between viral and bacterial infections. This is important because of increasing antibacterial resistance and could help doctors to tailor treatment quicker when someone is admitted with a suspected infection.

At present, distinguishing between different types of infections is complex and relies on symptoms, signs, a variety of clinical tests and clinical judgement. One of these tests is the CRP, which is used as an indicator of the severity of infection or inflammation, and is often used to monitor this over time. It is surprising that it has been used as one of the determinants in this new test, as it is considered to be a non-specific marker of inflammation or infection and increases in both viral and bacterial infections.

While the results of the study are positive, it’s important to realise that the test is not ready to be used on the general population. It will need to be tested on larger groups of people to confirm its accuracy. In addition, studies will need to show that it delivers benefits to patients in the way it is hoped – for example, finding out whether using this test leads to more accurate prescribing of antibiotics, less antibiotics being prescribed, or speeds up the process of diagnosing infection. Further research along these lines, including randomised controlled trials, will be required before it could be used in the clinical setting.

Although the test appeared to be good at distinguishing between viral and bacterial infections, it is unclear what results were obtained for people who did not end up with a clear diagnosis using the best existing methods. We do not know if the new test gave a result for these people or was inconclusive. This group doesn’t appear to benefit from the old or new testing methods, so will need to be explored in the next phases of research.

You can help slow down antibiotic resistance by always completing a course of prescribed antibiotics, even if you feel well before the end of the suggested course of treatment. Remember: antibiotics are not effective against colds, most sore throats and flu.

NHS Attribution