MONTREAL — New guidelines for the management of community-acquired pneumonia provide an excellent framework for site-of-care decisions, but they must be augmented with a good dose of clinical judgment, according to Glenn Tillotson, Ph.D.
The consensus guidelines from the Infectious Diseases Society of America and the American Thoracic Society hinge on two severity scoring systems to aid in the decision about whether to hospitalize patients with community-acquired pneumonia (CAP) or treat them as outpatients (CID 2007;44[suppl. 2]:S29).
However, although the well-established Pneumonia Severity Index (PSI) and the CURB-65 (confusion, uremia, respiratory rate, blood pressure, age 65 years or older) scoring systems are excellent tools, “clinical judgment based on more subjective criteria should override the rules,” Dr. Tillotson said at an international conference on community-acquired pneumonia.
Disease severity scoring systems adequately classify most patients with CAP into either hospital or outpatient treatment, but such systems may be less reliable for young adults (aged 17–40 years) and the frail elderly, said Dr. Tillotson, who is executive director of scientific affairs at Replidyne Inc.
He outlined one study of young CAP patients (median age 20 years) in which previous pulmonary disease, initial vital signs, and lab values were not predictive of mortality or length of stay (Chest 2006;130[suppl.]:105S). Another study found that in frail elderly patients with CAP, chronic comorbidities were not predictive of disease severity (Chest 2006;130[suppl.]:105S). “The factors we tend to lean on should not necessarily drive our decisions, especially in these populations,” he said.
Dr. Tillotson emphasized that although there may be a need for more specific scoring systems for CAP patients who are either young or frail and elderly, the science of scoring systems must always bow to the art of clinical judgment for all CAP patients.
“It's not always possible to articulate what marks a stay-at-home type patient versus someone who needs to go to hospital. There may be occasions when you should admit someone—not because they're severely ill, but simply because they need some TLC. One or 2 days in hospital could make all the difference,” he said.
He noted a study of almost 2,000 low-risk CAP patients in which 45% were treated as inpatients (J. Gen. Intern. Med. 2006;21:745–52). Among the hospitalized patients, about one-fifth had no identifiable risk factor according to the PSI. “An overnight stay in hospital can sometimes just get somebody over that hump. They're feeling sick; they have chest pains, fever, and coughing. They're not really sick enough to be admitted, but 24 hours in an observation ward can make a big difference,” he said.
Weighing against this idea are the known risks of hospitalization. “It's often better to manage someone in the community because they're less likely to acquire resistant pathogens, or they're less likely to have thromboembolic events,” Dr. Tillotson said.