Published in the September 2015 issue of Today’s Hospitalist
Editor’s note: This is the first in a two-part series on pneumonia care
WHILE MANY EXPERTS work to refine the concept of health care-associated pneumonia (HCAP) to curb the overuse of broad-spectrum antibiotic therapy, clinicians are trying to figure out how to de-escalate safely once it’s clear that pneumonia patients no longer need to be treated for drug-resistant organisms.
During his pneumonia presentation at this year’s Society of Hospital Medicine meeting, University of Michigan hospitalist program director Scott A. Flanders, MD, said that there is no randomized trial-generated evidence to help guide practice. But some observational reports show that de-escalation can be done safely for many HCAP patients. That includes patients whose blood or sputum cultures do and don’t grow organisms.
Dr. Flanders presented the findings of an observational, single-center retrospective chart review published in the October 2010 issue of Infection.
“It is hard to set an arbitrary threshold of two or three risk factors that requires broad spectrum treatment.”
~ Scott A. Flanders, MD, University of Michigan
That study looked at 102 patients at the University of California, San Diego, hospitalized with a diagnosis of HCAP. Two-thirds had their antibiotic therapy deescalated from the broad spectrum, triple-antibiotic regimen recommended by current HCAP guidelines to narrow-spectrum care (in this case, the respiratory fluoroquinolone moxifloxacin).
Nearly three-quarters of the patients had no organism grow on a blood or sputum culture, a group that researchers identified as “culture-negative.” That group as well as a “culture positive” group had similar lengths of stay, 30-day readmission rates and inpatient mortality. About three-quarters of the patients in both groups had their antibiotic therapy de-escalated.
De-escalation led to significant differences in outcomes that will impress most hospitalists. Patients whose therapy was de-escalated had shorter hospital stays (seven days vs. 13) and lower costs of care ($31,644 vs. $62,524). In addition, inpatient mortality among those de-escalated was 3%, compared to 28% in the non-de-escalated group.
Patients who remained on the triple therapy had more HCAP risk factors than patients who were de-escalated. HCAP risk factors include prior hospitalization, prior antibiotic use, tube feeding, non-ambulatory status, immune suppression, acid-suppressing therapy, dialysis and heart failure.
“The patients who didn’t get de-escalated looked sicker and probably didn’t get de-escalated because they were sicker,” Dr. Flanders said. Although these patients did not manifest greater “severity of illness as measured by the PSI” (pneumonia severity index), the authors wrote, the decision not to de-escalate these patients was probably due to “comorbidities that required prolonged hospitalization or other medical diagnoses that warranted continued use of broad-spectrum therapy despite negative cultures.” Median time to de-escalation was four days.
Dr. Flanders said he found this study “encouraging” because patients with negative and positive cultures were de-escalated safely, even though doctors had no guidelines to help them make those decisions.
Because most patients “never grow anything,” Dr. Flanders told hospitalists that they may be able to use the results of nasal swab PCR for methicillin-resistant staphylococcus aureus (MRSA) to help them make de-escalation decisions on HCAP patients. He referred hospitalists to two recent articles.
One, which was published in the February 2014 issue of Antimicrobial Agents and Chemotherapy, showed that a negative nasal swab was “pretty good at ruling out MRSA infection” and “can be reasonably used to guide antibiotic de-escalation.” On the other hand, Dr. Flanders noted, a positive swab did not mean the patient had MRSA. “Positive predictors weren’t good,” he explained.
A second study, published in the March 2013 issue of the same journal, showed that patients with negative nasal swabs could be safely de-escalated from MRSA-covering antibiotic regimens “in this case, vancomycin “when respiratory tract cultures weren’t available. According to the authors, “in-hospital mortality (7.7%) was similar to that of a previous study of de-escalation of antibiotics in pneumonia patients without adequate cultures.” They concluded that “in the absence of adequate lower-respiratory-tract cultures, it is reasonable to discontinue empirical vancomycin HCAP therapy in patients with negative MRSA nasal and throat cultures.”
As for which antimicrobials are best to de-escalate patients to, Dr. Flanders noted that there are no trials or guidelines. At the University of Michigan, the go-to regimen includes third-generation cephalosporins. Respiratory fluoroquinolones (levofloxacin or moxifloxacin) are also good choices, but he noted that it’s reasonable to “come up with something creative.”
Risks for drug resistance
Dr. Flanders recommended that hospitalists follow an algorithm for making de-escalation decisions. But rather than just assuming that all HCAP patients are the same, he said, doctors should start with determining how many HCAP risk factors patients have. An important study done by Japanese researchers and published in the Oct. 15, 2013, issue of the American Journal of Respiratory and Critical Care Medicine can help physicians decide which subset of health care-related pneumonia patients are more likely to have drug-resistant organism involvement.
“We are beginning to look at individual risk factors to see which ones are good and which ones aren’t” in terms of predicting patients who have a drug-resistant organism, Dr. Flanders said. The problem with the current HCAP guidelines? “Less than 10%” of these patients, he added, “ever have a resistant organism. There are things in the criteria that appear not to be strong predictors.” Examples, he said, include dialysis and wound care.
On the other hand, there are some “things that are critically important that are not reflected in the HCAP criteria,” such as “prior MRSA or pseudomonal infection.” Much work is going on, he said, to “figure out ways to risk stratify patients” and therefore predict who is less or more likely to need broad-spectrum antimicrobial therapy.
One risk stratification proposal, as laid out in the 2013 article from Japan, suggests adding up the risk factors, which are the same for community-acquired pneumonia (CAP) or HCAP. The more a patient has, “the higher the risk of having a resistant organism.”
Risk factors include prior hospitalization, immunosuppression, prior antibiotic use (in the last 90 days), gastric acid-suppressing therapy (either a proton pump inhibitor or H2 blocker), tube feeding and nonambulatory status. Based on that study, Dr. Flanders said patients with two or fewer risk factors “had very good outcomes” when treated following CAP guidelines. Those with four or more needed triple therapy.
“The three group was on the fence,” he added, “and you probably are going to cover broadly.”
Strong vs. weaker risk factors
An essential next step for this group of patients, he said, is to keep de-escalation in mind. “In these patients,” Dr. Flanders explained, “send sputum and blood cultures. If they are negative at 48 or 72 hours, de-escalate. That approach has been shown to be effective and reasonable.”
Some experts are trying to further refine the risk factors to predict which HCAP patients may need broad-spectrum antibiotics in case of multidrug-resistant organisms. In fact, Dr. Flanders said, there may be “strong risk factors” and weaker or unclear ones.
Factors that fall into the “strong” category include a prior hospitalization in the last 90 days; referrals from skilled nursing homes (rather than all nursing homes), especially if patients have been on antibiotics and have poor functional status; a history of MRSA or pseudomonas; and episodes of bronchiectasis or COPD with recurrent antibiotics and chronic steroids. “Weaker” risk factors include regular nursing home habitation, dialysis, wound care, home health care, feeding tubes and acid suppression therapy.
“It is hard to set an arbitrary threshold of two or three risk factors that requires broad spectrum treatment,” Dr. Flanders said. “I think you are going to adjust that cutoff based on how sick the patient looks and how prevalent resistant organisms are in your unit or hospital.”
If the patient looks sick and you have a lot of resistant organisms, “I think one risk factor should be enough for broad-spectrum treatment,” he added. “If your patient doesn’t look that sick and you have a low prevalence, maybe three or four risk factors are required to treat with broad-spectrum.”
Deborah Gesensway is a freelance writer who covers U.S. health care from Toronto.