Easing away from antibiotics for CAP
Physicians may want to consider holding off on prescribing antibiotics for lower-risk patients with viral pneumonia, according to a recently published American Thoracic Society guideline.
Physicians may want to consider holding off on prescribing antibiotics for lower-risk patients with viral pneumonia, according to a recently published American Thoracic Society (ATS) guideline, which also recommended lung ultrasound as a diagnostic alternative to chest X-ray.
The guideline on community-acquired pneumonia (CAP), which broadly focuses on reducing unnecessary antibiotic use, also suggested that physicians limit outpatient prescribing to three or four days in clinically stable patients who don't have underlying lung disease or other notable risk factors. That recommendation is shorter than the previously recommended regimen of five-plus days. The guideline was published on July 18 by the American Journal of Respiratory and Critical Care Medicine.
Each year, 1.2 million people seek ED care for pneumonia, according to federal data, and clinicians and researchers have advocated for strategies to decrease unnecessary antibiotic use. Even so, research reveals the gap between recommendations and practice. One retrospective analysis, which its authors described as the first North American study to look at appropriate antibiotic use for CAP in older adults, identified ongoing prescribing of longer regimens across two Canadian provinces. Seven days was the most common duration in the outpatient setting, followed by 10 days, according to the findings, which were published in 2023 in PLOS One.
The ATS guideline was informed by an emerging shift in the diagnosis and management of CAP, which stemmed in part from the broader use of lung ultrasound and rapid viral testing during the COVID-19 pandemic, said Barbara E. Jones, MD, MSCI, the committee's co-chair representing ATS.
The prior 2019 CAP guidelines, published by ATS and the Infectious Diseases Society of America (IDSA), had recommended prescribing antibiotics for patients with viral pneumonia in case they had an undetected bacterial infection, said Dr. Jones, a pulmonary and critical care physician and a physician investigator at the University of Utah and the VA Salt Lake City Healthcare System. But the escalated use of rapid viral testing for COVID-19 during the pandemic provided more insights, with physicians “identifying a lot more viruses than we ever did before,” she said.
“In the beginning of the pandemic, we were treating everyone with antibiotics,” Dr. Jones said. “By the end of the pandemic, we were becoming quite comfortable with holding antibiotics for a large proportion of patients with COVID-19, even the patients who were really sick.”
A prescribing shift
In sorting through the question of antibiotic prescribing for viral pneumonia, the committee members distinguished between patients with comorbidities and those without.
They recommended against physicians routinely starting antibiotics in patients without comorbidities who had been diagnosed with pneumonia based on clinical and imaging evidence and who tested positive for a virus. In these cases, committee members wrote, the low risk for an undesirable outcome if antibiotics are withheld or delayed “may not exceed the risks of harmful consequences of antibiotics to individual and public health.”
In contrast, patients with comorbidities face a higher risk of a serious outcome in the event of a bacterial co-infection, they wrote. Plus, that risk of co-infection may vary depending on the virus involved.
One 2021 retrospective analysis that the committee cited, published in the American Journal of Respiratory and Critical Care Medicine and involving ICU patients, found that nearly 34% of those with influenza also had a bacterial co-infection based on laboratory testing. But just 9.7% of patients hospitalized with SARS-CoV-2 tested positive for bacterial co-infection.
Guideline authors described these recommendations, along with others related to outpatient care, as conditional due to the underlying quality of the evidence. Given that no recommendation can account for all individual patient circumstances, they shouldn't be applied “by rote or in a blanket fashion,” they wrote.
The multidisciplinary committee included experts from the ATS and IDSA, but while the ATS board of directors approved the final guideline after multiple revision cycles, the IDSA board did not. Julio A. Ramirez, MD, FACP, the committee's co-chair representing IDSA, declined by email to discuss why that group didn't endorse the final version. IDSA will outline its rationale in an upcoming letter, he wrote. As of press time, the letter hadn't yet been published.
None of the studies reviewed as part of the guideline, apart from those involving lung ultrasound, were conducted during the COVID-19 pandemic, nor does the guideline apply to pneumonia treated while the pandemic was at its worst. The changing nature of the SARS-CoV-2 virus means that it's unclear whether patients today would benefit more from standard CAP management or from treatments used during the pandemic, the authors wrote.
The new guideline does reflect one of the pandemic lessons that “there's a lot of viral pneumonia out there, and it doesn't necessarily need to be treated with antibiotics,” said Michael B. Rothberg, MD, MPH, an internal medicine physician and vice chair for research at the Primary Care Institute of the Cleveland Clinic.
“People with viral pneumonias are unlikely to have a co-infection, especially if they're well enough to come to the doctor's office,” said Dr. Rothberg, who published an In the Clinic overview about CAP in Annals of Internal Medicine in 2022. “If they don't look like they need to go to the ED right now, they probably don't have a bacterial co-infection. It's OK to not treat them right away.”
For an older patient with a serious medical condition, such as chronic obstructive pulmonary disease (COPD) or heart failure, the calculus is different, Dr. Rothberg said. “Even if it's a small probability of a bacterial infection, the harm associated with missing it is much greater.”
Sorting through comorbidities
The ATS guideline didn't identify any outcomes-related studies involving antibiotic prescribing for viral pneumonia, instead relying on other evidence for the recommendation, including clinical experience, noted Bryan Kraft, MD, an associate professor of medicine at the Washington University School of Medicine in St. Louis.
That new recommendation may prove challenging for physicians to adopt, given that typically they prescribe antibiotics to patients in their office when, for instance, they test positive for influenza and infiltrates are visible on a chest X-ray, said Dr. Kraft, who also coauthored a 2022 review article in Frontiers in Medicine looking at practical antibiotic stewardship in the outpatient setting.
“It is good to give outpatient clinicians permission to watch and wait rather than automatically prescribe an antibiotic,” Dr. Kraft said. “But to some it may be uncomfortable, or disquieting, to do that.”
The committee members differed to some extent regarding which comorbidities pose sufficient risk that they may tip the balance in favor of antibiotics.
The highest percentage of committee members, 65% or more, agreed that patients with end-stage liver and renal disease or COPD (other than asthma) fell into that more vulnerable category, according to a table listing 17 medical conditions. Chronic kidney disease, HIV (CD4 cell count >200 mm3), asthma, rheumatological disease (but not treated with immunosuppressants), and obesity ranked at the bottom, with 24% or fewer of the committee members in agreement.
The guideline details other potential variables, including the patient's history of pneumonia and their personal preferences, in weighing whether to hold off on prescribing. “Some patients may not want to take antibiotics. Some patients may be fearful not to take them,” Dr. Kraft said. “This [guideline] emphasizes an individualized approach to patients, which will require a conversation about risks and benefits with each patient. That could be challenging in a fast-paced office setting.”
A diagnostic alternative
During the early days of the pandemic, physicians relied more on point-of-care ultrasound (POCUS) of the lungs, amid the risks involved in transporting patients to radiology, said Nilam Soni, MD, MS, FACP, a guideline committee member and ultrasound expert who is also a professor of medicine and an academic hospitalist at the University of Texas School of Medicine in San Antonio. “It did, for a lot of reasons, put point-of-care ultrasound into the spotlight,” he said.
With sufficient training, primary care physicians can make more informed decisions about whether to start antibiotics in patients with symptoms of CAP by using POCUS if they don't have easy access to a chest X-ray, Dr. Soni said. “Why do we use [POCUS]? Because it improves our diagnostic accuracy at the bedside so we can sleep better at night,” he said.
Lung ultrasound demonstrated a median sensitivity of 95% and specificity of 75% in patients with suspected CAP, according to a meta-analysis of 11 studies involving 939 patients, which was published in the guideline. The median sensitivity of chest X-ray was 70%; the specificity was 55%.
The committee conditionally recommended that lung ultrasound “is an acceptable diagnostic alternative to chest X-ray in medical centers where appropriate clinical expertise exists.”
That's a key caveat, Dr. Soni said. “The bias of these studies is that they're mostly done by people that use ultrasound, and they're mostly people who are enthusiastic to embrace the technology.”
Dr. Kraft made a similar point. “The ultrasound can lead you astray if you don't know what you're looking at,” he said. “Achieving competency requires hours of training, and hundreds of exams, as well as ongoing use to maintain skills. Ultrasound also would not be able to distinguish between viral or bacterial pneumonia.”
Increasingly, younger physicians are honing their POCUS skills during residency, but primary care physicians already in practice may not be able to make that time commitment, Dr. Kraft added. For those who do, lung ultrasound can expand their practice skills, enabling them to obtain an immediate thoracic image.
“If a patient presents with shortness of breath, POCUS could be very valuable in a patient's initial evaluation to hone the differential diagnosis,” Dr. Kraft said, such as heart failure, pleural effusion, or pneumonia. (ACP offers a POCUS pathway for internal medicine physicians.)
Tightening up regimens
The 2019 ATS/IDSA guidelines recommended five to seven days of antibiotics for outpatients whose conditions are improving, and the more recent ATS guideline ratchets that time frame down even further in that population.
Committee members suggested five days of antibiotics, with a minimum of three, unless the patient exhibits signs of more severe pneumonia or has underlying lung difficulties, or faces hurdles in getting follow-up care. The shorter duration applies only to patients whose blood pressure, temperature, oxygen levels, and other vital signs indicate clinical stability.
This recommendation is based on research indicating that mortality rates aren't any higher in stable patients who receive fewer than five days of antibiotics than in those who get longer regimens, Dr. Jones said. Meanwhile, the shorter courses for patients who improve quickly don't just benefit public health by guarding against further antibiotic resistance in the community but also potentially protect individuals. Along with reducing the risk of diarrhea, yeast infections, and other side effects, ongoing research points to the importance of not unnecessarily disrupting the microbiome, she said.
ACP's best practice advice for antibiotic prescribing, published in Annals of Internal Medicine in 2021, also mirrors the ongoing shift toward shorter regimens. The CAP portion of the advice, based in part on the 2019 ATS/IDSA guidelines, recommended a minimum of five days in patients who show signs of clinical stability.
In younger healthy patients, an antibiotic regimen as short as three days “can be appropriate,” said Rachael Lee, MD, MSPH, FACP, lead author on ACP's advice, professor of medicine in the division of infectious diseases at the University of Alabama at Birmingham, and a member of ACP's Population Health and Medical Science Committee.
But physicians should consider prescribing for five days if the patient has multiple comorbidities or other lung conditions, such as COPD, or signs of severe pneumonia on chest X-ray, Dr. Lee said. Regardless of duration, careful antibiotic selection is important, with related guidance provided in the 2019 ATS/IDSA guidelines, she said. The recent ATS guideline, she noted, also includes the caution that azithromycin and clarithromycin may not be adequate for outpatient treatment due to the rising rates of macrolide-resistant Streptococcus pneumoniae in the United States.
Above all, physicians should ensure that patients have a clear follow-up plan in case their symptoms worsen, both Dr. Lee and Dr. Kraft stressed. A nurse or another clinician may call to check in. Or, if a physician prescribes a delayed prescription, they should make sure that the patient understands when symptoms warrant filling it, Dr. Kraft said.
“The less aggressive the antibiotic prescribing is, the more aggressive the monitoring and contingency plans should be,” he said. “So, if there are barriers to follow-up, for instance, you may want to treat more aggressively.”
Dr. Rothberg similarly emphasized a follow-up plan, including ensuring that patients understand which symptoms warrant seeking further care. “They should look for worsening breathing—that's really the main thing. Or ‘Are you feeling faint? Do you feel sicker? Is your fever going up?’”
Patients may not necessarily feel better the day after starting antibiotics but shouldn't feel worse and should be reminded of that accordingly, Dr. Rothberg said: “‘If you feel worse the day after I've given you antibiotics, I need to see you again.’”
In general, the ATS guideline represents a “growing appreciation of both the unmeasured harms of antibiotics and also the fact that antibiotics, the way that we gave them traditionally, are often not necessary,” Dr. Rothberg said.
Still, he agreed with Dr. Kraft that the shift to prescribing fewer than five days in lower-risk patients may initially be difficult for internal medicine physicians.
“You need to have doctors try it, and get experience with it, and see that it's OK,” he said. “A lot of patients are getting five days now. They used to get 10 days. The fact that we went from 10 days down to five days is a huge victory.”
