Michigan COVID-19 hospital patients got unneeded antibiotics during surge

More than half of Michiganders hospitalized for coronavirus during the first several months of the pandemic were unnecessarily given antibiotics. That is in part because testing delays meant doctors didn’t know whether patients had COVID-19, another potentially dangerous infection like strep or pneumonia, or both.  

While antibiotics don’t treat COVID, they can increase the risk that a patient will develop a resistance to antibiotics later on, when the treatments may be desperately needed, said Dr. Valerie Vaughn. She’s an assistant professor and hospitalist at the University of Michigan, and one of the authors of a new study published in the journal of Clinical Infectious Diseases.  

• The latest: Michigan coronavirus unemployment, map, curve, updated COVID-19 news
• Dashboard: Michigan coronavirus testing numbers, trends, COVID-19 data

Researchers looked at 1,705 randomly selected patients admitted at 38 Michigan hospitals between mid-March and mid-June, during the state’s surge. 

While nearly 57 percent of those patients were given antibiotics early on, the vast majority turned out not to have needed them: only 3.5 percent ended up actually having a bacterial infection at the beginning of their hospital stay. 

“There have been lots of studies that have shown that people who die from COVID often have bacterial infections,” Vaughn said. “Now, whether those infections actually caused their deaths, we don't know for sure. But we know that [COVID] patients that have a bacterial infection, about half of them will die.”

Which is why receiving unnecessary antibiotics is so risky for COVID patients, Vaughn said. 

“People with COVID are often in the hospital for long periods of time. They're getting devices placed that increase the risk of infection, or being exposed to those superbugs in the hospital. So suddenly now it's three, four weeks down the line. They're still in the hospital, and now they actually do have an infection. Well, the fact that they got those antibiotics in the beginning makes it harder to treat that infection down the line.”

COVID patients were also more likely to get unnecessary antibiotics if they went to a for-profit hospital, researchers found, or “if they were older, had a lower body mass index, had more severe disease (e.g., respiratory support, severe sepsis),” or if X-rays showed an infection in the lungs.

What doctors knew, or didn’t 

Part of the problem is that some COVID symptoms can mimic bacterial infections, like pneumonia. And at the beginning of the pandemic, while testing was taking longer, doctors may have prescribed antibiotics - just to be safe - while waiting for test results. 

“These patients are coming in with fevers, with abnormal chest X-rays, with [a] cough, they're requiring oxygen supplementation,” said Dr. Joel Fishbain, the medical director for infection prevention and hospital epidemiology at Beaumont Grosse Pointe. 

Bacterial coinfections were also more common in previous pandemics, Vaughn said, which is why doctors may have been concerned about them during the initial months of the new coronavirus. Since then, however, research has emerged showing coinfections are less common in COVID patients early on in their hospital stays.

“Once we knew for sure that patients had COVID, and COVID was what was responsible for their cough, their shortness of breath, their high fevers, then we could feel more confident with stopping their antibiotic,” Vaughn said.

In fact, more than half the antibacterial treatment given to patients occurred before COVID tests results turned positive, researchers found. Once doctors knew it was COVID, 54 percent stopped prescribing antibiotics within a day of getting the results. And as the pandemic progressed and turnaround time for tests decreased, so did the antibiotic use. 

“So it is this piece of: The quicker you can get a test result back, the quicker you can either prevent antibiotic use or stop it once it's already been started,” Vaughn said. 

When patients get more antibiotics 

Among patients who kept receiving antibiotics even after a positive COVID test, only 7 percent actually had a confirmed bacterial coinfection. Of those who didn’t have a coinfection, 36 percent stayed in the hospital for fewer than 5 days; 39 percent stayed between 5-7 days; and 24 percent stayed longer than a week. 

Dr. Fishbain said  he wouldn’t be overly concerned about patients who may have received just a single, initial dose of antibiotics before their test results came back. 

“One dose is more risky than no doses, but one dose is much, much less risky than five doses,” he said. 

“So the standard treatment would be five doses of an antimicrobial agent for community-acquired pneumonia, except for Legionella. So five days is our typical treatment course. And five days carries a lot of potential risk of developing resistance (to the antibiotic), versus one dose. So it’s sort of like an escalating risk: the more antibiotics, you give more resistance.”

While it’s too early to say whether COVID patients who received unnecessary antibiotics were more likely to become severely ill or die, Vaughn said it’s possible. 

“As a physician who treated a lot of these patients, I can tell you that firsthand, I saw a lot of patients who went on later in the hospitalization to develop bacterial infections that were really hard to treat [and] made patients really sick,” she said. “And many of them passed away from it.

“There was also another side effect of antibiotics related to this really bad diarrhea called C. difficile infection. And those infections are caused by antibiotics, are made worse by antibiotics and can actually be quite deadly. And so I saw quite a few patients with these C. difficile infections.”

That’s where Vaughn is hoping future research can shed some light. 

“That's part of the next step of this work. And hopefully we'll be able to combine our data with other groups as well to look and show those outcomes. But I can tell you that as a doctor, I saw them firsthand.”

With antibiotics, like so many aspects of COVID, doctors have had to learn on the fly.  

“When we started in March, there was very little knowledge about how to take care of [these] patients,” Vaughn said. “And what we saw was really rapid improvement over the following months. We've seen that not just for antibiotics, but for other treatment around COVID. So I actually think we're going to keep moving in that direction because people really want to do the right things by their patient.”