Mitigating antibiotic resistance in respiratory infections: an expert roundtable

As a result of increasing antibiotic resistance, clinicians are often in the challenging position of acting in the best interest of each patient while trying to avoid contributing to the problem with the unnecessary or inappropriate use of antibiotics. However, many studies have shown that delaying antibiotic administration for serious infection in the emergency department (ED) can have devastating consequences.

In a 2018 observational cohort study of 117 patients admitted to the intensive care unit (ICU) with sepsis or septic shock, there was an increase in mortality risk for each hour of delay in ordering (22%) or administering (15%) antibiotics after triage.¹ These delays were also associated with a greater number of days spent in the hospital and the ICU. In line with these findings, the authors of a 2017 systematic review examined 14 studies regarding antibiotic administration in patients with sepsis in the ED, and found the greatest mortality benefit in those individuals who received appropriate antibiotic treatment within 1 hour of recognition.²

Pulmonology Advisorinterviewed the following experts to discuss considerations pertaining to antibiotic therapy when treating patients in the ICU for pneumonia or tuberculosis: Keira A. Cohen, MD, assistant professor of medicine in the Division of Pulmonary and Critical Care Medicine at Johns Hopkins University School of Medicine in Baltimore, Maryland; Stanley Deresinski, MD, clinical professor of medicine in the Division of Infectious Diseases and Geographic Medicine at Stanford University Medical Center in California and medical director of the Stanford Antimicrobial Stewardship Program; and Angela Rogers, MD, assistant professor of medicine in the Division of Pulmonary and Critical Care at Stanford.

Pulmonology Advisor: What are the risks of delaying antibiotics for pneumonia? For tuberculosis (TB)? How are these situations typically approached in the ICU?

Dr Cohen: The issue of TB in the ICU is quite uncommon, as the overwhelming majority of patients with TB do not require ICU care. With people who require ICU-level care for infection, we are usually talking about sepsis, and we know that the time to initiation of appropriate treatment is very important to optimize treatment outcomes. If someone requires ICU-level care for pneumonia because of a high blood culture, septic shock, or respiratory failure, appropriate antibiotics are very important, as mortality increases with each hour that antibiotic treatment is delayed.

Everyone wants early, appropriate treatment, but we often do not know what the appropriate treatment is when the patient is initially admitted to the ICU. So, we choose empiric antibiotic therapies based on guidelines from professional societies and each hospital’s local resistance patterns. If we choose increasingly broad therapies, we could overtreat and expose patients to certain toxicities. And there are antibiotics that we need to reserve for known drug-resistant organisms, such as ceftazidime/avibactam, so that we do not introduce antibiotic resistance into the community.

Dr Deresinki: Most cases of pneumonia acquired by individuals living in the community are caused by viruses or bacteria. For treatable causes, early initiation of antibiotic therapy is preferred to avoid progression of the infection. The only viral cause of pneumonia that is currently treatable is that resulting from influenza, for which oseltamivir is indicated. A second drug effective against influenza virus infection, baloxavir, has recently also become available. Cases of pneumonia acquired in the community can be treated with a number of different antibiotics.

Dr Rogers: In both bacterial pneumonia and TB, you are balancing the benefits of treatment right away and avoiding the disease getting worse vs exposing patients to unnecessary or potentially harmful drugs. With bacterial pneumonia, the risk is mostly for the patient who is being treated: their condition could get worse. I work mainly in the ICU, and with patients who are already sick where there is no room for error or waiting if they get worse, so we have a very low threshold for starting empiric antibiotic treatment. 

With TB treatment, certainly there is the risk to the patient for delaying, but there is also a big societal risk: TB is often infectious, and while you are waiting for results to come back, there is the potential that individuals with TB may be out in society exposing others.

Pulmonology Advisor: What are some key considerations in choosing the right antibiotics for pneumonia or TB?

Dr Cohen: TB is changing: we have very different diagnostic tests now. I will say that that sometimes the antibiotic choices we make can decrease the likelihood of detecting TB; for example, the use of fluoroquinolones can decrease our ability to diagnose TB. These are second-line agents for TB, so treating bacterial pneumonia with a fluoroquinolone could reduce your diagnostic yield if you are looking for TB later.

Dr Deresinki: The problem for the clinician is that, with some exceptions, it may not be possible to distinguish between viral and bacterial pneumonia without the use of rapid diagnostic tests, which may not be readily available in all clinical settings.

Dr Rogers: Bacterial pneumonia can become deadly very quickly, and so we tend to treat it empirically. If we can get a sputum or blood culture that tells us the specific bacteria involved, that is a huge help and allows for more targeted treatment, but that is often not possible. So, a lot of times, you base your treatment regimen on the history, appearance on chest X-ray, the patient’s own risk factors and baseline health, and knowledge of the common bacterial resistance patterns in your hospital or area.

With TB, it is usually, although not always, a more indolent course, meaning that people have had it for weeks or months rather than days. That gives you a chance to make sure you have the diagnosis right, often with multiple sputums or invasive bronchoscopy for a sample if you cannot make the diagnosis otherwise. Because TB takes a while to culture, it can take some time to get resistance patterns, so you still need to choose a regimen empirically based on your knowledge of whether TB in your area or where the patient traveled is sensitive to standard therapy. So, you still try to treat quickly, both for the patient and to limit any exposure to others.

TB is quite rare in most regions of the United States. At a pulmonary review course 2 years ago, people raised their hands if they had ever seen a case of TB, and less than half of the room had seen it. That means it is not on clinicians’ radars, so we need to keep a high index of suspicion.

Pulmonology Advisor: In considering whether to delay antibiotics in patients with TB or pneumonia, and in choosing which ones to give, are there more problems with certain classes?

Dr Cohen: There is no debate about whether antibiotics should be delayed — they should always be given immediately. But in terms of which antibiotics to give, you have to think about where the patient came from. For example, if they have not been in and out of hospitals often, they are less likely to have been exposed to resistant bacteria. If a patient is doing poorly, the clinician has to broaden the antibiotics used and adjust course based on how the patient responds.

Dr Deresinki: TB may occasionally mimic other more common bacterial pneumonias, but more often presents with more prolonged symptoms. Treatment, which differs from that for other causes of pneumonia, is not usually initiated until the diagnosis is confirmed. Unfortunately, the diagnosis is often not initially considered, and antibiotics are administered, and one of the antibiotic types that may be administered are fluoroquinolones, such as levofloxacin. Because the fluoroquinolones are partly effective against Mycobacterium tuberculosis, their administration may actually delay the ability to diagnose this infection.

When the diagnosis of TB is confirmed, treatment is initiated immediately not only to improve the patient’s clinical status but also to reduce the risk of transmission to others. Treatment may have to be altered when the infection is caused by drug-resistant strains.

Dr Rogers:In general, on a patient-by-patient basis, I do not think this is such an issue. On a societal level, a lot of patients with viral illness receive treatment with antibiotics, which has no benefit, exposes them to risk, and can increase the risk for antibiotic resistance to the entire community. On an individual level, we know that treating with antibiotics, even if you do not experience drug-related adverse effects, wipes out your good bacteria and puts patients at risk for things such as Clostridium difficile diarrhea, so each decision to treat with antibiotics should not be taken lightly.

Pulmonology Advisor: Are there any additional points you would like to mention?

Dr Cohen: It is very important to provide empiric antibiotic therapy in a timely manner. There are now diagnostic tests, such as blood-based assays, that help us more rapidly diagnose infections in the ICU. In general, we usually give patients about 48 hours of empiric broad-spectrum antibiotic coverage. When patients come into the ICU with sepsis, we want additional culture data to direct antibiotic treatment most appropriately. It is also important to obtain those data via blood, sputum, and urine testing, etc.

References

1. Kim RY, Ng AM, Persaud AK, et al. Antibiotic timing and outcomes in sepsis. Am J Med Sci. 2018;355(6):524-529.

2. Sherwin R, Winters ME, Vilke GM, Wardi G. Does early and appropriate antibiotic administration improve mortality in emergency department patients with severe sepsis or septic shock? J Emerg Med. 2017;53(4):588-595.

Source: Pulmonology Advisor