Antibiotics may carry more cost than benefit for patients at end of life

Experts offer advice on discussing and deciding when to use or avoid.


Prescribing an antibiotic for a terminally ill patient with a suspected infection may be viewed as an obvious step in care, even when more invasive measures to prolong life are no longer being considered.

Certainly, use of antibiotics is common in such situations. Ninety percent of hospitalized patients with advanced cancer receive antimicrobials during the week before their deaths, as do 25% of hospice patients in their final weeks and 42% of nursing home residents with advanced dementia in their last 2 weeks of life, according to a viewpoint in the Nov. 17, 2015, Journal of the American Medical Association (JAMA).

Photo by Thinkstock
Photo by Thinkstock

Similarly, 21.1% of adult inpatients discharged to hospice care from Oregon Health & Science University over a 3-year period were given an antibiotic prescription at discharge, according to a study in the September 2014 Antimicrobial Agents and Chemotherapy. Patients who were being discharged to home hospice care and those with a cancer diagnosis had a significantly greater chance of receiving a prescription.

Infections are prevalent at the end of life, and it's difficult to prove that these examples of antibiotic use are overuse, but Jon P. Furuno, PhD, associate professor at Oregon State University College of Pharmacy and one of the investigators on the Antimicrobial Agents and Chemotherapy study, believes they are.

The drugs may provide little to no benefit for patients whose life expectancy is counted in days or weeks, according to him and other experts. “We don't know what [the antibiotics] are doing, how they work, or if they are benefiting patients with respect to their symptoms,” said Dr. Furuno.

Benefits and risks

Antibiotics for terminally ill patients can have some benefits. Infections can be painful, and treatment may provide some relief, particularly for urinary tract infections, but the data supporting that assumption “are not particularly strong,” Dr. Furuno said. An antibiotic can possibly also ease the discomfort of a bad case of oral thrush, noted Paul Sax, MD, an infectious disease specialist and professor of medicine at Harvard Medical School in Boston.

There is also the potential to prolong life, at least by a little. “There is some evidence to support that people will live longer but will have a higher symptom burden and lower quality of life when their infections are treated,” Dr. Furuno said. That tradeoff might be worth it for a patient who wants a few days to say goodbye, attend a daughter's wedding, or get his affairs in order, for example.

On the other hand, the same goals might be achievable by other means. “The general assumption is that the risks of an antibiotic at the end of life are quite low, so just give it. But there are some data to suggest that the best palliative care may be better than antibiotics alone at relieving some symptoms associated with infection,” said Manisha Juthani-Mehta, MD, FACP, an author of the viewpoint in JAMA and associate professor of medicine and infectious diseases fellowship program director at Yale School of Medicine in New Haven, Conn.

Even if the risks of antibiotic use are low, they can be significant to frail patients. Opportunistic infections, including Clostridium difficile, can be both acutely unpleasant and fatal. Intravenous antibiotics infused with a volume of saline can lead to fluid overload. Drug interactions can occur and complicate care.

Antibiotics can also have nervous system side effects, such as agitation and confusion, and there is evidence indicating that some patients who receive antibiotics for pneumonia have a worse quality of life than those not treated, said Dr. Sax. For patients at the end of life, “putting them on an antibiotic to help their pneumonia get better is really of questionable long-term benefit,” Dr. Sax said.

Antibiotics can also harm patients by prolonging the dying process, noted Leah Rosenberg, MD, attending physician in the division of palliative care at Massachusetts General Hospital in Boston.

There's also the risk to others, pointed out Julia M. Gallagher, MD, medical director of the Massachusetts General Hospital Home-Based Palliative Care Program. She recommends “carefully weighing the risk for the emergence of resistant organisms and the potential to transmit those organisms from patient to patient against the possibility that there may be little to no benefit to a particular patient given where that patient is in the trajectory of his or her illness.”

Perhaps based on these concerns, more than 60% of physicians who responded to a survey said they would not want antibiotics at the end of their own lives, Dr. Sax reported in a Feb. 15, 2015, blog post for the New England Journal of Medicine. “Those recurrent urinary tract infections, aspiration pneumonias, infected pressure sores, and other indignities of our failing bodies can be treated with antibiotics (though with progressively less effect)—but to what end?” Dr. Sax wrote.

Patient perspective

Of course, the patient's goals of care should help guide any decision about antibiotic use, the experts noted. Many patients may have made their wishes known about aggressive interventions such as intubation or dialysis in an advance directive but may not have thought about or been given an opportunity to consider whether they want an antibiotic at the end of life.

Some families and patients may feel strongly that antibiotics should be prescribed, even if an infection is not confirmed by lab testing and other heroic end-of-life measures have been discontinued.

“It is very easy for them to say, yes, keep [my family member] on an antibiotic. At that point they are almost thinking in some way that they don't want to abandon everything,” Dr. Sax said. Other families and patients may feel that the patient has suffered enough, and they do not want the suffering prolonged with antibiotics. Some may want both palliative care and antibiotics.

The only way to know is to ask. “Addressing the issue of antibiotics is important and [physicians] should not assume that prescribing or continuing an antibiotic is a given,” Dr. Juthani said.

A physician may be able to suggest a middle road for patients and families having difficulty with the decision. Lenny Noronha, MD, a hospitalist at the University of New Mexico in Albuquerque, said that he will often offer an antibiotic if there is hope that someone could recover and their goal is not yet comfort care.

“I will say that we can try this for 24 to 48 hours and then we will reassess regularly whether it is helping or harming. I will explain that, depending on the type of infection, people tend to respond in this amount of time, but if we are not seeing improvement by that time, we will discuss whether or not to continue,” said Dr. Noronha.

If the antibiotic doesn't work and needs to be withdrawn, patients and their families should be part of that conversation, too. “If the family member at that point insists that the loved one stay on the antibiotic, then that is OK. It is more important for that family member to be able to feel like they did everything they needed to do for their loved one,” Dr. Noronha said.

Whether the antibiotic is stopped or never started, patients and families need support and reassurance that the patient has not been abandoned. They should be assured that the health care team “will continue to remain engaged and provide any and all interventions that best ensure the comfort and dignity of the patient,” said Dr. Gallagher.

Dr. Juthani described this task in an article published Jan. 7, 2015, in Next Avenue, “As an infectious disease physician, my most important role may not be to advise which antibiotic to use or which diagnostic test to perform next, but to sit with a distressed family to help justify that it is OK to let go,” she wrote.