Pop quiz: Do you know the guidelines for treating health care-associated pneumonia (HCAP)? And is your clinical practice in accordance with those guidelines?
If the answer to either of those questions is no, you're not alone. A recent study, published in Clinical Infectious Diseases last December, presented 1,300 physicians at academic medical centers with clinical scenarios involving hospitalized pneumonia patients. The results highlighted the general lack of knowledge surrounding this relatively newly defined disease.
The physicians, who included hospitalists, internists and emergency and critical care specialists, were asked to select the appropriate antibiotic regimen for a pneumonia patient. In six of the nine scenarios, the hypothetical patient met guideline criteria for HCAP, but almost 80% of the doctors failed to select guideline-concordant therapy. Only 1% of the group would have prescribed appropriate therapy in all six HCAP scenarios.
The results were not entirely surprising to study co-author Bradley A. Sharpe, FACP, an associate clinical professor in the department of medicine at University of California, San Francisco.
“There are some sets of guidelines that people would be familiar with, but the HCAP guideline was relatively obscure when it came out,” he said. That was in 2005, and the guidelines, developed by the American Thoracic Society (ATS) and Infectious Diseases Society of America (IDSA), called for antibiotic coverage of methicillin-resistant Staphylococcus aureus and resistant Pseudomonas in all at-risk patients, which includes residents of skilled nursing facilities, patients on hemodialysis and the recently hospitalized, among others.
The study participants claimed familiarity with the recommendations, however. More than 70% of the respondents said that they were aware of and practiced according to the HCAP guidelines. Pride is one possible explanation for this discrepancy. “No one feels comfortable saying they've never heard of national guidelines. Even though it's an anonymous survey, people want to feel and show that they are updated on the literature,” said Dr. Sharpe.
The participants could also have been telling the truth about their practice patterns, if not their knowledge, suggested lead author Gregory B. Seymann, ACP Member, an associate professor of health sciences at the University of California, San Diego. “They may be practicing in environments where they use an order set, either computerized or paper, where they check off antibiotics so they don't have to remember these things. That would actually be a good thing.”
Another explanation is that, regardless of whether they are familiar with them, the study participants disagree with the guidelines. A number of experts, including one of the guidelines authors, do. Based on recent research, anecdotal experiences and even local hospital culture, physicians have developed their own strategies for treating HCAP—not all of which are concordant with the guidelines.
The source of debate and uncertainty is the use of broad-spectrum antibiotics—the drugs that effectively combat MRSA and Pseudomonas. The guidelines recommend an initial regimen of those medications in all patients who meet the criteria for HCAP, a strategy that can seem like overkill in some cases.
“We see a 50-year-old patient who's on dialysis, who has pneumonia, who's on a regular floor bed. I don't think that person routinely needs broad-spectrum gram-negative and MRSA coverage. They can probably be treated with coverage for routine community-acquired pneumonia,” said Dr. Sharpe.
Such a relatively healthy patient wasn't what the experts were envisioning when they originally developed the guidelines, acknowledged Michael S. Niederman, FACP, a guideline author and chairman of the department of medicine at Winthrop-University Hospital in Mineola, N.Y. He noted that the ATS/IDSA guideline group is working on plans to revise the HCAP recommendations.
“Our original concept of HCAP was probably much too narrow. What we were doing in the guidelines was probably thinking about the more severe end of the spectrum of patients with HCAP—people who come to the hospital, people who often end up in the ICU, people with multiple risk factors,” he said.
Much of the data that support broad-spectrum treatment were gathered among such severely ill patients. “The studies of HCAP in which patients had resistant organisms were all or mostly patients that had positive cultures,” said Dr. Sharpe. “We know that the sicker patients are more likely to have positive cultures.”
On the other hand, the results of HCAP studies also show that not giving broad-spectrum drugs to the patients who need them can have serious consequences. “If we don't treat patients with the right antibiotic up front, their risk of death is greater and their costs of care are greater,” said Marin H. Kollef, FACP, author of an editorial that accompanied the guideline-compliance study and a professor of pulmonary and critical care medicine at Washington University in St. Louis.
It's not a decision that can be reversed, either, researchers have found. “The literature shows that patients with HCAP who have their antibiotic regimen escalated after the cultures come back positive die at the same rate as patients whose antibiotics are left alone,” reported Dr. Seymann. “If the patient is one of those who have Pseudomonas or MRSA, you don't want to wait to cover that for a few days.”
The potential consequences of missing drug-resistant pneumonia are a strong argument for following the guidelines, but there are also compelling reasons not to use these drugs when they may not be needed.
“There's a big effort to reduce antibiotic overuse in hospitalized patients. If you go by the letter of the law of the guidelines, any patient with one of these risk factors ends up on three drugs, vancomycin being one of those. We have major problems with overuse of drugs like vancomycin,” said Scott A. Flanders, FACP, director of the hospitalist division at the University of Michigan Health System, Ann Arbor.
“[Physicians] want to try to get by with the narrowest agent they can in order to try to preserve antibiotics for the long term,” agreed Dr. Niederman.
In addition to concerns about resistance, cost may be encouraging physicians and, to an even greater degree, hospitals to minimize use of the broad-spectrum drugs. Some hospitals limit hospitalists' ability to prescribe these antibiotics, said Andrew F. Shorr, ACP Member, associate director of pulmonary critical care at Washington Hospital Center in Washington, D.C.
“Hospitals labor under the concept that by restricting antibiotics, they'll either save cost or prevent resistance, if they restrict them to infectious disease [specialists] only. The problem with that is that the infectious disease provider isn't on the front line seeing any of these patients,” Dr. Shorr said.
In addition to the potential negative impact on outcomes, such restrictions may not even lower costs. “Your pharmacy budget may look smaller, but what happens is you have patients spending longer in the hospital because you're playing catch-up,” Dr. Shorr said. “In fact, when you get it right the first time, all the studies show that you actually use less total antibiotics because the patients get better faster.”
The pneumonia patients who get better fast present another challenge to potential prescribers of the broad-spectrum—and usually intravenous—drugs. “What do you do on hospital day 3 when the patient is ready to go home? Do they need intravenous vancomycin at home even though they got better?” asked Dr. Sharpe.
Taking patients off the heavy-duty drugs is exactly what some experts and the guidelines would recommend. De-escalation of antibiotics allows physicians to balance concerns about covering drug-resistant bacteria and avoiding overuse of antibiotics, Dr. Kollef said. “If we get the cultures before antibiotics are started, then after 24 to 48 hours we can look at the patient. Now we should be able to de-escalate the antibiotic regimen,” he explained.
Dr. Flanders has a similar de-escalation effort at his hospital. “We have a fairly aggressive approach to try to have everybody at 48 to 72 hours reassess all these antibiotics. If now you don't think the patient has HCAP, if they're doing a lot better than you expect, if you haven't cultured a particularly nasty organism, we really focus on narrowing those antibiotics down.”
Such a strategy is not always easy to implement, however, according to Dr. Sharpe. “Your average clinician may say, ‘They're better on this therapy. How can I stop any of these things, even though cultures are negative?’ We have trouble convincing both housestaff and faculty to stop vancomycin when there's no evidence of MRSA,” he said.
Even if physicians want to treat based on cultures, the specimens are more difficult to gather from patients who aren't intubated, said Dr. Seymann, citing a September 2004 Archives of Internal Medicine study in which a good-quality sputum culture was available in only about 14% of pneumonia patients. Without accurate cultures, the decision to de-escalate has to be made on fuzzier criteria. “In a patient who has improved with broad-spectrum therapy, how can one distinguish improvement as a result of antibiotics from improvement despite antibiotics?” Dr. Seymann asked in a May 2008 article in the Journal of Respiratory Diseases.
Finding a plan
Determining the cause of patients' illness without definitive lab data is the main challenge of HCAP treatment. To meet it, physicians on the front lines have evolved individualized, but not dissimilar, strategies.
“People base it on a lot of gestalt. How does the patient look? The sicker they are, the more willing doctors are to give very broad-spectrum treatment,” said Dr. Flanders.
Dr. Sharpe also uses and teaches severity of illness as a determinant of therapy. “If they're sick—sick defined as ICU, respiratory failure, sepsis—those patients get very broad-spectrum antibiotics. But if they're not ill and they're simply getting admitted to a regular acute care bed, then maybe we start with run-of-the-mill CAP therapy and see how they do.”
Severity is one of several risk factors recommended for consideration in a 2008 consensus paper on HCAP, published in Clinical Infectious Diseases. “What we tried to do is say not every patient needs to be treated for drug-resistant pathogens with a broad spectrum, just patients with multiple risk factors. The multiple risk factors include severe pneumonia, immunosuppression, prior antibiotics, recent hospital stay and poor functional status,” explained Dr. Niederman, an author of the paper. According to the paper's algorithm, broad-spectrum drugs should be given to patients with severe pneumonia and at least one other risk factor, or to nonsevere cases with two other risk factors.
But that algorithm may need revision if future research indicates that some of the included risk factors are better predictors than others, Dr. Niederman warned. “Are all risk factors equal or do some have more importance than others? There are some studies that suggest they may not all be equal. That's not been well worked out.”
Considerations may also vary depending on where you practice. “Everybody's got a different problem. In my institution, our biggest resistant pathogen that comes in the door is MRSA, so understanding the burden of MRSA is a lot more important to us than understanding Pseudomonas. The whole way to approach this is to look at your local data,” said Dr. Shorr.
Whether it's based on gathered data or just the opinion of neighborhood experts, Dr. Sharpe suspects that many physicians are following a local HCAP plan instead of the national guidelines. “You develop a local culture and decide ‘this is how we treat Disorder X.’ My guess is that if you presented clinicians at different institutions with a nursing home patient with pneumonia, the different places would treat them differently.”
Technology, in the form of computerized order sets, may be the best tool for eliminating this variation and standardizing care, Drs. Seymann and Sharpe suggested. “If you have guidelines and want to implement them, just sending them to people in a PDF or giving them a lecture is unlikely to change behavior,” said Dr. Sharpe.
“Even myself, when I'm admitting someone, I don't have to remember all the different seven or so risk factors for HCAP. It prompts me in the computer so I get it right,” Dr. Seymann said.
Technology is also advancing HCAP diagnostic tests. “Some are currently available and being developed that will very rapidly—within a period of hours—tell us what the organism is. Then you will be able to get that antibiotic therapy refined within six hours of the patient entering the hospital,” said Dr. Kollef.
Once physicians have access to better diagnostics, revised guidelines, and more evidence (data are need from randomized or at the least prospective studies of HCAP, experts said), Medicare reporting might be the next step to improving care, Dr. Sharpe predicted. Currently, Medicare does not have a treatment standard for antibiotic choice in HCAP.
“When you're compared to the hospital down the street and your antibiotic compliance rates are incredibly low, that motivates people to change. That certainly has improved compliance with guidelines for CAP,” he said. “A publically reported therapy for HCAP—that's 10 years away, but in the future.”
In the meantime, hospitalists are somewhat on their own in making these difficult decisions. “They should evaluate each patient individually,” said Dr. Niederman.
“Have a high index of suspicion and recognize that not all pneumonia that presents to the emergency room is exactly the same,” advised Dr. Shorr.