Have you ever been told you have MRSA or VRE?

You are sitting in a hospital triage area. The nurse looks up from her clipboard: “Have you ever been told you have MRSA or VRE?” She says it like it is routine. You have never heard those letters in your life. 

After my last post, a few readers asked questions that made me realize I had skipped a step. Before I get into how we interpret resistance data at scale, it helps to step back because the question that nurse just asked is connected to a problem that touches all of us. 

MRSA and VRE are bacteria that have evolved resistance to important antibiotics. You can carry these bacteria without feeling sick, sometimes for months after an infection clears. The hospital asks whether you have ever been diagnosed with either for two reasons: if you are still carrying the organism, they need to protect other patients from it and if you develop an infection during your stay, your doctors need to know your history so they do not start you on antibiotics that have already failed. 

But this is not just a hospital problem. Antibiotic resistance affects anyone who might ever need an antibiotic, which is most of us. A urinary tract infection that used to clear up in days might now need two or three rounds of different antibiotics before one works. A child’s ear infection might not respond to the first-line antibiotic, which is the initial preferred treatment for specific bacterial infections. Even routine surgery carries more risk when post-operative infections become harder to treat. The antibiotics we have taken for granted our entire lives are becoming less reliable and it is happening quietly enough that most people do not notice until it affects them directly. 

When every line on the report says R 

When someone develops an infection, doctors can order what is called a culture and sensitivity report. A sample is taken and sent to the lab, where the bacteria are grown and then exposed to a range of antibiotics. The result is a report: each antibiotic listed next to a letter. S means susceptible—the antibiotic should work. R means resistant—it will not. You want to see a lot of S’s on that report. 

 In 2016, a woman in her seventies was admitted to a hospital in Reno, Nevada, with a serious infection following a bone injury. She had spent months in hospitals abroad. Her report came back and every row read the same thing: R. She was resistant to all 26 antibiotics her doctors could try, including colistin, the antibiotics doctors reach for when absolutely nothing else works. There was no next option. Unfortunately, she developed septic shock and died weeks later.  

Her case made headlines. But the underlying problem is not rare. Antibiotic-resistant infections directly kill over 1 million people every year, and that number is projected to nearly double by 2050. Meanwhile, the pipeline for new antibiotics is not keeping pace, often taking over a decade from discovery to bedside use. We are losing antibiotics to resistant bacteria faster than we are replacing them. 

Where does the DNA come into this? 

The culture and sensitivity report—grow the bacteria, expose them to antibiotics, see what works—remains the gold standard for guiding treatment, and for good reason. But it has limits. Growing bacteria takes time, sometimes days and not all organisms can be cultured in a lab at all. More fundamentally, resistance is an evolutionary problem. It is driven by changes in bacterial DNA and there are questions about those changes that a lab report was never designed to answer. That is why scientists began decoding the genetic makeup of bacteria at large scale—tens of thousands, even hundreds of thousands of organisms. With that kind of information, it becomes possible to identify which resistance genes are present, trace where they originated, understand what they mean for the person who is sick and help guide treatment and prevention decisions closer to real time. 

But right now, that genetic information is piling up far faster than anyone can work through it. The custom analyses needed to interpret each case are slow and resource-heavy and the promise of near-real-time insight from all this data is not yet being met. The ways researchers are working to close that gap—building on the interpretation layers I described last time—is what I will be writing about in my next blog post. 

Back in triage, you answered “no” to the nurse’s question. She ticked a box and moved on. But now you know what those letters mean and what is at stake when the answer changes. 

Image by Miguel Á. Padriñán from Pixabay