Evidence-Based Practice in LTC

Know Pneumonia

The reality versus the evidence of diagnosis & treatment

by Steven Levenson, MD, CMD
Multi-Facility Medical Director, Baltimore, MD
Chair, Caring's Editorial Board

Charles Crecelius, MD, PhD, CMD
Past President, Missouri Association of Long-Term Care Physicians
Medical Director, Delmar Gardens, St. Louis, MO
Member, Caring's Editorial Board

About This Series: Over the years, certain practices have become common in nursing home care nationwide. Consultants, management, and surveyors offer multiple opinions and conflicting advice that may confuse staff and practitioners and result in over- or under-treatment. Various sources, including the Centers for Medicare and Medicaid Services, are advising nursing home staff to refer to "best practices." Yet only some of these practices reflect reliable evidence. Other habitual approaches are problematic and should change. Often, "best practices" is erroneously interpreted to mean that there is a single best treatment, when in reality there may be "better" and "worse" practices. This series will review common practices in nursing homes, comparing habitual to desired approaches, and may recommend important changes.

A common clinical issue in nursing homes (NH), pneumonia can be difficult to accurately diagnose and treat. Lack of knowledge by nurses and physicians may lead to pneumonia under-recognition or (more commonly) over-diagnosis and subsequent excess use of resources and antibiotics.

Common Practices

NH care providers may not appreciate the symptoms of pneumonia. Reliance on fever and cough may discourage recognition of pneumonia. Lab tests and chest X-rays may be used too often to judge the severity of a patient's illness or to determine whether to prescribe antibiotics.

If staff members misunderstand the limits and benefits of pneumonia treatments, including hospitalization, then inappropriate expectations and use of resources may result. Also, undue fear of litigation may compel nurses and physicians to order additional tests, use antibiotics prematurely, and send patients to the hospital needlessly. Financial realities, such as higher reimbursement per visit and more visits allowed, may encourage physicians to hospitalize patients.

Many nurses and physicians are tempted to over-treat upper respiratory infections. Residents and families may feel the patients should receive antibiotics with even mild symptoms in a misguided attempt to avoid serious illness. Often, physicians prescribe antibiotics due to time constraints and the pressure to treat instead of obtaining a thorough patient assessment. This can result in potential complications related to indiscriminate medication use (i.e., antibiotic-associated colitis) and, ultimately, excess morbidity and mortality.

The Evidence

The incidence of pneumonia is higher in NHs than in any other setting, ranging from 0.3 to 2.5 episodes per 1,000 resident days. Risk factors may include poor functional status, dysphagia or the presence of a feeding tube, malnutrition, increasing age, use of sedating medications, and acute confusional state (delirium). Aspiration pneumonia is less common than other kinds of NH pneumonia. It results more commonly from aspiration of oropharyngeal matter than gastric contents--neither of which is usually witnessed.

Streptococcal pneumonia is the most common pneumonia pathogen, followed by Haemophilus influenzae and Moraxella catarrhalis. Gram-negative bacilli are isolated in 0 to 12% of residents with pneumonia, with atypical organisms isolated even less frequently. In most NH communities, streptococcal species have shown increasing resistance to macrolide antibiotics. NH residents and staff should be vaccinated against influenza and streptococcal pneumonia.

Identifying pneumonia requires a high index of suspicion, as symptoms may be atypical. Of NH patients with pneumonia, cough may present in only about 60%, fever in about 65%, and altered mental status in about 60%. Tachypnea (respiratory rate >25) presents in about 40% of NH patients and may be the most sensitive early indicator of pneumonia. Universal agreement doesn't exist regarding the exact criteria for diagnosing pneumonia. A practical definition of pneumonia is a new infiltrate on chest X-ray with at least two clinical signs (new/increased cough or sputum production, fever >38° C, pleuritic chest pain, new relevant chest exam findings, and dyspnea/tachypnea, or worsened mental/functional status).

Physician examination alone is not sensitive (47% to 69%) enough or specific (55 to 75%) enough to consistently diagnose pneumonia. Examination by either a physician or a physician extender hasn't been shown to improve mortality (although all authorities suggest prompt evaluation). Frequent re-evaluation by nursing staff during the first three days of new onset pneumonia is highly recommended empirically.

Most pneumonia practice guidelines suggest obtaining both a white blood cell count with a differential and a chest X-ray to substantiate a diagnosis. Patients with multiple symptoms likely have pneumonia, so don't delay treatment while waiting for results. Pulse oximetry is warranted for NH residents with dyspnea or tachypnea, but the results shouldn't be used alone to determine severity or need for hospitalization. While helpful, sputum culture proves difficult to obtain in NH patients. Blood cultures have a low yield and are not routinely suggested. Assess and manage the patient's hydration status and consider other likely infections.

Historically, two-thirds to three-fourths of NH patients with pneumonia are treated without hospitalization. Hospitalizing patients has not been shown to help maintain function or improve survival rates. In the presence of renal insufficiency, tachycardia, malnutrition, functional dependency, and mood deterioration, patient mortality is higher--regardless of where they are treated. No definitive guidance exists to help physicians choose treatment settings. Criteria for hospitalization may include degree of instability of vital signs, presence of active co-morbidities, the NH's capabilities to monitor and manage the patient, and resident/family directives.

Current guidance promotes quinolones as the first-choice antibiotic for NH-acquired pneumonia; second choice is amoxicillin-clavulanate (or parenteral second/third generation cephalosporin) plus a macrolide. Streptococcal species may be increasingly resistant to quinolones. Use of parenteral antibiotics hasn't been shown to be better than oral antibiotics in affecting patient mortality. Duration of treatment is empirical; most guidelines suggest 10 to 14 days of therapy; however, longer treatment may increase the risk of complications, such as antibiotic-associated colitis.

For patients with advanced dementia, pneumonia treatment remains controversial. Existing decision-making guidelines may apply to fewer than half of relevant patients. Review your advance directives and surrogate instructions. Regardless of whether treatment is rendered, your patients should at least receive supportive measures to diminish their pain and relieve discomfort.

Conclusions

A common infection, pneumonia has variable presentations in NH patients. Careful assessment of clinical symptoms, coupled with chest X-ray findings and selected lab tests, can usually establish a diagnosis. The utility of tests or interventions, such as physician assessment and pulse oximetry, remains limited. Evaluate patients sufficiently to distinguish upper- and lower-respiratory infections because the treatments differ.

Many suggested treatment guidelines remain empiric and warrant further research. Administer antibiotics promptly if related signs and symptoms suggest a clinical diagnosis of pneumonia. Because they appear to be as effective as parenteral antibiotics, oral antibiotics should be used more often--especially to treat mild to moderate infections. Using oral antibiotics can save the cost, discomfort, nursing time, risks, and inconvenience associated with intravenous therapies.

The role of hospitalization for NH patients with pneumonia is unclear and does not appear to have a significant effect on morbidity or mortality. Therefore, facilities should employ simple measures, such as improved monitoring and provision of general patient support, to avoid the risks and patient discomfort associated with hospitalization.

The opinions expressed in this column are those of the authors and do not necessarily reflect those of the American Medical Directors Association.

back to top