Content written by Dr. Clayton Mowrer.
Early in my medical training, my father developed a rapidly-progressive type of cancer. Over the course of several months, his health declined quite quickly, and he was spending more time in the hospital than outside of it. It became uncomfortably clear that pushing forward with invasive and toxic interventions would cause him much more harm and discomfort than it would any benefit. Our family had many gut-wrenching discussions with my dad’s physicians, and we eventually decided that transitioning him to hospice care would be in his best interest in order to focus on keeping him comfortable in his final days of life. Though that time is mostly a blur to me, I do recall that we were able to discharge him from the hospital and he was able to die peacefully, surrounded by family, without the hassles, tests, and complications that come with being in the hospital.
The discussions surrounding end of life (generally defined as the final days or weeks of life) are incredibly difficult. While the goals of palliative and hospice care being that primarily of minimizing suffering and maximizing quality of life, it is not always a clear or easy decision as to which interventions fall under this definition. Most tend to agree that interventions such as CPR, intubation, or medications with high risk for side effects – e.g. chemotherapy – are not in line with the goals of comfort care and are therefore commonly avoided in end of life care; conversely, the approach to the treatment of infections and the use of antibiotics at the end of life is a much more polarizing topic.
And the discussion of the use of antibiotics is particularly important, as many studies and observations have shown that many patients at the end of life have a high risk of infections due to a weakened immune system, comorbidities, very frequent exposure to healthcare facilities (where infections are easily spread), and adverse effects of medications such as chemotherapy. Consequently, antibiotics are often given in these final days or weeks of life.
However, here is not much known about the true prevalence of infection at the end of life, with research showing that antibiotics are commonly prescribed in terminally ill patients in the absence of clinical evidence of bacterial infection, due in part to the view that antibiotics are historically viewed as relatively benign. But, just as with the general population, antibiotics don’t come without their own risks, and it is important to thoroughly understand the risks and benefits of antibiotic therapy in order to have an informed conversation when the decision to move towards comfort care is made.
Antibiotics, to be sure, can be quite beneficial if used in the appropriate clinical scenario. Particularly when there is a proven bacterial infection, an appropriate course of antibiotics can provide relief of pain associated with the infection, especially in infections such as urinary tract infections that can cause significant discomfort. Additionally, some patients may have certain events that are important to their quality of life – such as a wedding or graduation – which they would like to attend, and there is evidence that the treatment of a documented infection in terminally ill patients may prolong life just a little bit.
Yet, suspected infections are not often proven, leading to the frequent and long-term use of broad-spectrum antibiotics (known as empiric antibiotics). Such broad-spectrum antibiotics come with risks, including liver and kidney toxicities, as well as an increased risk for developing Clostridiodes difficile infection (C diff) – an infection that can lead to profound diarrhea, resulting in intensified distress. The administration of antibiotics itself can carry some risk if necessary to give intravenously: IV’s can cause irritation to the skin and soft tissue, occasionally leading to further/additional local or disseminated infections. In the setting of patients who may be exhibiting delirium or an altered mental status, restraints could be necessary.
Patients, their families, and providers should also be aware and take into consideration what the evaluation of a suspected infection entails. Hospitalization, with many blood tests and imaging, is typically involved. In addition – though this is not always a concern for patients and their families, I do believe it is important to consider – there can be a financial burden that accompanies pursuit of infectious diagnostic workup and treatment. Workup of suspected infections are a frequent cause for hospitalizations in terminally ill patients and can lead to prolonged stays and numerous diagnostic tests, which can be costly.
Finally, studies have shown that greater antibiotic use at the end-of-life is associated with the acquisition of multidrug-resistant organisms. Addressing these organisms has become a priority in the field of medicine worldwide and has been specifically targeted by organizations such as the CDC, who released a recent report regarding the threat of antibiotic-resistant organisms in the United States.
Patients in end-of-life care and their families, with their medical providers, should include antibiotic use in discussions of goals of care, as, though it can have some benefits, it also carries distinct risk for harms and should be considered in a similar manner as other treatment interventions. In this way, the comfort of the patient can remain the ultimate focus.
I encourage the reader to read the writings of Timothy Sullivan and Manisha Juthani-Mehta MD (below), on this topic, as they are much more eloquent than I.
- HIV and ID Observations, by Paul Sax, MD
- Antibiotics Are Often Used at the End of Life, But At What Cost? By Timothy Sullivan
- Why Infection May Be a Good Way to Die
- Infect Dis Clin N Am. 2017 Dec; 31(4): 639–647. https://doi.org/10.1016/j.idc.2017.07.009
- JAMA. 2015 Nov 17; 314(19): 2017–2018. https://doi:10.1001/jama.2015.13080
- Cancers (Basel). 2016 Sep; 8(9): 84.
- Chest. 2010 Sep; 138(3): 588-594. https://doi.org/10.1378/chest.09-2757
- Journal of Pain and Symptom Management. 2003 May; 25(5): 438-443.https://doi.org/10.1016/S0885-3924(03)00040-X
- Journal of Pain and Symptom Management. 2000 Nov; 20(5): 326-334. https://doi.org/10.1016/S0885-3924(00)00189-5
- Journal of Pain and Symptom Management. 2013 Oct: 46(4): 483-490. https://doi.org/10.1016/j.jpainsymman.2012.09.010