In this issue of American Family Physician, Mills and colleagues review the treatment of patients with nursing home–acquired pneumonia¹ and discuss the recent guideline² on hospitalized adults with pneumonia, published by the American Thoracic Society (ATS) and the Infectious Diseases Society of America (IDSA). The underlying issue in the treatment of nursing home–acquired pneumonia is the debate regarding the etiology of this infection. Among nursing home residents with pneumonia who are treated in the nursing home or the hospital setting and who do not require intensive care, the etiology is similar to that of community-acquired pneumonia (CAP)^3,4; Streptococcus pneumoniae and Haemophilus influenzae are the most common causative agents of nursing home–acquired pneumonia. For many years, however, the prevailing dogma has been that gram-negative aerobic bacilli and Staphylococcus aureus, including methicillin-resistant S. aureus (MRSA), are important causes of nursing home–acquired pneumonia, especially in persons hospitalized for treatment.⁵ This dogma is based, in part, on studies of nursing home–acquired pneumonia, published in the past three decades, that used sputum cultures to define etiology. However, these studies have been criticized because investigators who reported gram-negative bacilli and S. aureus as common causes of nursing home–acquired pneumonia did not evaluate the quality of the sputum sample that was cultured.⁴ Studies that used sputum culture results only if the sample was not contaminated with oropharyngeal flora found gram-negative bacilli and S. aureus to be uncommon causes of nursing home–acquired pneumonia.⁶

The notion that gram-negative bacilli and MRSA are common causes of nursing home–acquired pneumonia has also influenced the classification of this infection. In the past, nursing home–acquired pneumonia has been included in studies of CAP. Two recent guidelines on the management of CAP included specific recommendations for treatment of patients with nursing home–acquired pneumonia in the hospital setting.^7,8 These guidelines did not consider resistant gram-negative bacilli or MRSA to be major etiologic agents of nursing home–acquired pneumonia. However, the recently published ATS/IDSA treatment guideline for hospital-acquired and ventilator- and health care–associated pneumonia placed nursing home–acquired pneumonia in the health care–associated category for the first time.⁸

Health care–associated pneumonia includes infection in a heterogeneous group of patients: those receiving home wound therapy, antibiotic therapy, chemotherapy, or dialysis; those who have recently been hospitalized; and those who reside in nursing homes. The common denominator among these patients is the supposed high risk of pneumonia caused by resistant gram-negative bacilli or MRSA. The rationale for including nursing home–acquired pneumonia in this group is based on the results of two studies that found a high rate of isolation of gram-negative aerobic bacilli and MRSA among nursing home residents with pneumonia who required mechanical ventilation.^9,10 This latter group represents only a small proportion of the nursing home residents who have pneumonia. Furthermore, in one of these studies, no bacterial pathogen was isolated in 43 percent of episodes, despite the use of bronchoscopy and quantitative bacteriology.¹⁰

Several potential explanations for this finding have been proposed: previous antibiotic therapy, a viral infection (culture for viruses was not done), or a noninfectious cause of acute respiratory failure including acute aspiration (gastric content) pneumonitis.¹¹ Thus, even in this supposedly high-risk group, gram-negative aerobic bacilli and MRSA were not pervasive.

Classifying nursing home–acquired pneumonia as a health care–associated infection raises several concerns. First, which guideline takes precedence in terms of treatment recommendations for hospitalized residents with pneumonia? The recommendations in the ATS/IDSA guideline² do not match those published in other CAP guidelines for treating nursing home–acquired pneumonia in hospitalized residents^7,8; this difference in recommendations is not discussed in the ATS/IDSA guideline.² Second, because the ATS/IDSA guideline² may be interpreted as replacing previous guideline recommendations,^7,8 physicians might try to apply this guideline to all nursing home residents hospitalized with a diagnosis of pneumonia. Third, the ATS/IDSA guideline² may be promoted as a standard of care by which to judge physician or hospital performance in the management of nursing home–acquired pneumonia, even though there are no clinical trials to support the validity of the recommendations.

It is highly unlikely that randomized, controlled trials on nursing home–acquired pneumonia will be conducted to determine the most appropriate treatment regimens in the hospital or nursing home setting. There is no impetus on the part of pharmaceutical companies or the federal government to fund studies in frail nursing home residents with pneumonia. On one hand, there has been considerable concern about increasing antibiotic resistance and promoting the judicious use of antibiotics. At the same time, experts are recommending regimens that are broad in antibacterial spectrum for treating nursing home–acquired pneumonia with minimal evidence that this is necessary, other than in residents with severe pneumonia on mechanical ventilation (who represent less than 5 percent of all residents with pneumonia).²

New methods (e.g., molecular techniques) are needed to accurately and rapidly identify the etiology of nursing home–acquired pneumonia. Until such methods are available, antibiotic regimens will tend to be based on dogma. This means that frail, debilitated nursing home residents with pneumonia who are admitted to the hospital may be exposed to potent antibiotics despite minimal evidence that such treatment is required. This also places these patients at risk of complications, including promoting colonization with resistant organisms, and adverse drug reactions. Given the current evidence, I would suggest that it is still appropriate to consider previously published guidelines.^7,8 Until additional data become available, empiric treatment with antibacterial agents intended to cover resistant gram-negative bacilli and MRSA should be reserved for those who have acute respiratory failure requiring mechanical ventilation, or for those who have other risk factors for infection with resistant organisms (e.g., hospitalization in the previous 30 days, antibiotic therapy in the previous 30 days, a history of colonization or infection with a resistant organism). For now, physicians should have flexibility in making decisions regarding the treatment of hospitalized nursing home residents with pneumonia who are managed in the non–intensive care setting. We should not be required to follow a guideline that does not necessarily apply to our individual clinical settings.