The general goal of antimicrobial therapy is to increase patient's survival and cure infectious diseases. Because antimicrobial agents are unable to alter the progression of underlying diseases or rather sometimes can prolong suffering in patients at the end-of-life, the goal of prolonging life should be weighed against the risk of prolonging suffering [18]. Thus, the goal of antimicrobial therapy shifts from improving survival to symptom control with a shift in the goal of care from cure to palliation [19]. The goal of antimicrobial therapy for terminal patients at an early stage can include symptom control and improvement in survival while maintaining the quality of life. The goal of antimicrobial therapy for terminal patients at the end-of-life is achieving an improvement in symptoms. Antimicrobial agents led to symptomatic relief in only 33% of patients with advanced cancer during the last several weeks of life and in an even smaller proportion of patients (9.2%) during their last week of life [20]. Although symptom improvement was hardly observed when antimicrobial agents were administered during the last week of life, antimicrobial use was common among patients with advanced cancer even after transitioning to comfort care [21]. The risks and burdens of antimicrobial therapy should be carefully examined when comfort is the intended goal of care [21].

Polypharmacy is common in terminally ill patients owing to multiple comorbidities and increase adverse drug reactions, drug interactions, and costs. These potential risks are weighted when antimicrobial agents are administered. In addition, intravenous antimicrobial therapy, which is commonly performed using intravenous devices in acute in-hospital care settings, is associated with the risks of phlebitis, local skin, and soft tissue infections as well as secondary bacteremia [18]. Clostridioides difficile infection, which is common and often recurrent, is another important potential harm associated with antimicrobial use [22]. Antimicrobial use in end-of-life care is considered as a risk factor for MDRO colonization that adds psychological burden through infection control measures including patient isolation and contact precautions which conflict with the goals of palliative care [51923]. When antimicrobial therapy against infections caused by MRDOs is indicated according the discussions for goals of care, the available treatment options are often limited, invasive, expensive, or associated with adverse effects that distress patients and their families [19]. All these potential harms of antimicrobial therapy can make the quality of life poor or shorten the survival duration in terminally ill patients with multiple comorbidities.

The diagnosis of infection in a palliative patient is difficult. Common indicators for infections such as fever, leukocytosis, and elevated acute phase protein level are not always present in patients who have multiple comorbidities and are taking medications that obscures an infection. In addition, many end-of-life patients with cognitive impairment have difficulty in exposing their symptoms. On the contrary, fever, leukocytosis, and elevated acute phase proteins are present in non-infectious conditions such as malignancy. Medications as well as malignancy and pulmonary embolism/deep vein thrombosis can cause fever occasionally. Procalcitonin plays a role in the diagnosis of infections in cancer patients [24].

A cross-sectional study on attending physicians and fellows with appointments at either the University of Pennsylvania or Children's Hospital of Philadelphia conducted in 2017 revealed that approximately half of the physicians believed that antimicrobial use in end-of-life care contributes to antimicrobial resistance and approximately 30% of physicians intended to continue antimicrobial administration for the patients even after care-giving was deemed medically futile or when comfort was elected to be the main goal of care [25]. Nearly one-fifth of pediatricians intended to continue antimicrobial therapy for hospice patients whose death was imminent compared to few adult physicians (2.7%; P <0.001) [25]. The reasons why the physicians may continue antimicrobials at the end of life includes meeting family expectations, wanting to avoid the perception of “giving up,” uncertainty about prognosis, and reducing patient pain or discomfort [25]. A meta-analysis that investigated symptom improvement with antimicrobial therapy in end-of-life patients revealed marked variability, i.e., 60 – 92% of patients with urinary tract infections and 0 – 53% with respiratory infections showed an improvement [26]. Despite the lack of compelling evidence, many physicians may continue antimicrobial therapy to relieve patient symptoms because they perceive adverse consequences to be at a relatively low frequency [25]. However, the relationship between antimicrobial administration and quality of life improvement remains controversial [27]. A survey conducted in Korea in 2014 to explore opinions about end-of-life discussions among oncologists and residents revealed that antimicrobial therapy for end-of-life care was discussed only by approximately 17% of the respondents [28].

Among 255 American patients who had advanced cancer and had been taking community-based outpatient hospice and palliative care in 2003, a majority of them (79.2%) chose either no antimicrobial therapy (31%) or symptomatic use only (48.2%) [29]. Other studies performed in Taiwan, which has a distinctive culture deeply influenced by Confucianism and Buddhism, have revealed that 45.8% of the patients with terminal cancer expressed their wish to use antimicrobials even in the very terminal stage and 26.4% did not wish to use antimicrobials [7]. The most influential persons were medical professionals [7]. A study on 3,840 Korean individuals (1,242 cancer patients, 1,289 family caregivers, and 303 oncologists) showed that a large majority of individuals supported the withdrawal of futile life-sustaining treatment (87.1 – 94.0%) and use of active pain control (89.0 – 98.4%) [30]. A smaller majority (60.8 - 76.0%) supported withholding of life-sustaining treatment [30]. Another study on the discussion regarding end-of-life care among Korean elderly patients and their family caregivers through focus group interviews reported a wide range of diverse attitudes and opinions across all study questions and suggested that the end-of-life care discussion should focus upon devising an approach which takes into account the individualized needs of these elderly patients and their family members [31].

Infection is an important and normal part of dying process for many patients, and thus often cannot (or should not) be “treated” [32]. In a retrospective chart review of all deaths from 1998 to 2000 in an inpatient hospice unit in San Antonio, Texas, pneumonia was found to be present in 79% of patients at autopsy and appeared to be the major cause of death in 44% of patients [33]. Infections and febrile episodes are the most common acute complications experienced by terminally ill patients. It is difficult to distinguish between the normal part of the dying process and the aspects that are either reversible or clinically reasonable to treat. The question of whether to withdraw or withhold antimicrobials at the end-of-life cannot be definitely answered based on scientific evidence. The decisions for antimicrobial therapy at the end-of-life are complicated and multidimensional that includes ambiguity of therapeutic effectiveness, physician perception, quality of life, attitudes and opinions of patients and family caregivers, ethics, and public health.