It isn’t always the coronavirus — or any virus — that kills patients with COVID-19. Bacteria can, too. While bacterial and viral infections manifest as similar symptoms, they are treated very differently, as antibiotics cannot kill viruses. A common complication of viral infections such as the flu or the coronavirus is a secondary, superimposed bacterial infection — or a superinfection — resistant to most antibiotics, including ones used earlier during treatment.

This complication might not get much discussion outside hospitals, but it doesn’t surprise infectious-disease experts. Decades after the Spanish flu pandemic of 1918, several studies showed that many deaths were caused by secondary bacterial infections.

Although the limits of laboratory testing a century ago and the lack of complete records make pinpointing exact causes of death difficult, experts generally attribute a lot of the deaths to complex pneumonia on top of the Spanish flu. While medical understanding of the internal damage caused by COVID-19 today is still evolving, there is growing concern about the potential risk of superinfections for the sickest of patients.

“Up to 50% of the deaths of people hospitalized on ventilators so far are actually a consequence of bacterial superinfections,” said Julie Gerberding, Merck and Co. chief patient officer and former director of the U.S. Centers for Disease Control and Prevention. “It’s a force of mortality that we need to pay more attention to.”

COVID-19 weakens patients to the point that they are less able to fight off other infections — an issue that can be compounded by the hospital treatments meant to save them.

“They often have not only ventilation tubes but other catheters and devices that set the stage for hospital bacteria to start infections,” Gerberding said.

Furthermore, COVID-19 patients tend to be in the hospital for a long time, which means a greater chance of transmission of bacteria. Antibiotics are commonly prescribed to prevent patients from developing secondary infections, or infections that occur during or after treatment for another infection, but they don’t always work. Resistant strains have become a crisis in recent decades as overuse and improper use of these medicines have allowed mutations to flourish.

“Superinfections can be caused by antibiotic resistant bacteria, making them very difficult to treat. Not all hospitals have the antibiotics readily available, and because we don’t have a robust pipeline of new antibiotics, we are steadily running out of treatment options,” Gerberding said. “Just when you thought it was about the virus, there’s more to it.”

There’s another complication to all this: Recent evidence from two mouse studies published in Science suggests that a potential therapy being evaluated in clinical trials for COVID-19 — interferons — may increase susceptibility to bacterial infections. Interferons are a type of molecule naturally released by the body, with specific types designed to kill viruses. Scientists have studied their potential for years as a type of antiviral treatment. Of course, mouse trials don’t necessarily correlate directly to human health, but the phenomenon is concerning.

“There are three families of interferons. Type 1 and Type 3 are known for their antiviral properties,” said Ivan Zanoni, assistant professor of pediatrics at Harvard Medical School and senior investigator of one of the studies published in Science.

Type 3 interferons have attracted interest as a potential treatment for COVID-19 because compared with Type 1 interferons, they are less potent but longer acting.

“Our research shows that a prolonged exposure to Type 3 interferons can cause a problem to the host,” Zanoni said.

The specific problem is damage to the lungs. Two studies in mice showed that a prolonged exposure to Type 3 interferons can reduce lung cells’ ability to recover. These epithelial cells serve as important gatekeepers that prevent the lungs from being infected by bacteria.

“Any breakdown of this epithelial barrier is going to predispose someone to bacterial superinfections,” said Alison Carey, associate professor of pediatrics and immunology at Drexel University College of Medicine. “These studies emphasize the delicate balance between the need to kill the virus and prevent the long-term negative associations.”

But finding the exact timing and duration of treatment can be a challenge.

“The main problem is you don’t know when your patient was infected. You have a vague idea of when their symptoms started, and you ideally need to treat people as early as you can,” said Andreas Wack, group leader at the Francis Crick Institute and senior author of the other interferon study published in Science. “But later in the infection is when the repair of the lungs needs to happen, and when there are still interferons sticking around, this can be a problem.”

Experts emphasize that there still is a great deal to learn about the long-term effects of COVID-19 and even treatments used against it.

“Even when we find a vaccine, the COVID story will not end because there may be long-term effects — potentially increases in cases of COPD [chronic obstructive pulmonary disease] and other antibiotic-resistant bacterial infections,” Wack said.

“There’s a lot that is unknown about what the long-term consequences are going to be for the COVID-19 pandemic,” Carey said. “We can look at other respiratory infections to give us an idea of what we have ahead of us, but in terms of prognosis and long-term complications, we don’t know yet.”