Aspiration Pneumonia: Where Have All The Anaerobes Gone?

Abstract & Commentary

Synopsis: Anaerobes were not found to play an important role in the etiology of pneumonia in nursing home residents with an increased risk of aspiration.

Source: El-Solh AA, et al. Microbiology of severe aspiration pneumonia in institutionalized elderly. Am J Respir Crit Care Med. 2003;167:1650-1654.

El-Solh and colleagues in Buffalo obtained respiratory specimens by protected brush alveolar lavage from 95 elderly residents of long-term care facilities admitted to the hospital with severe aspiration pneumonia. All 95 had at least 1 comorbid condition, with 78% having had a previous cerebral vascular infarct. None had radiographic evidence of necrotizing pneumonia or cavity formation. Cultures were processed within 30 minutes of acquisition. At least 1 microorganism was recovered from 54 (57%) of the patients. Anaerobic bacteria were recovered from 11 (20%) of those with positive cultures and aerobes from 43 (80%). Six of the 11 from whom anaerobes were recovered had mixed anaerobic-aerobic infection.

Among the 67 pathogens isolated, 49% were Gram-negative enterics, 16% were anaerobes, and 12% were Staphylococcus aureus. The most frequently isolated Gram-aerobic, Gram-negative bacilli were, in decreasing order of frequency, E coli, K pneumoniae, Serratia spp., and P mirabilis; there were only 2 P aeruginosa and 1 Enterobacter cloacae isolated. In addition to S aureus, Streptococci, including S pneumoniae, comprised the Gram-positive aerobic bacteria isolated. Six of the 11 anaerobes isolated were Prevotella spp., 3 were Fusobacterium spp., and 1 each were Bacteroides spp. and Peptostreptococcus spp.

Anaerobic bacteria were more likely to be recovered from patients with significantly impaired functional status. There was not a correlation, however, between a history of prior episodes, and the presence of a percutaneous gastrostomy tube or the presence or degree of periodontal disease.

Thirty-five patients died, including 24 with progressive pneumonia and 11 with multiorgan system failure. Whether anaerobes were identified or whether antibiotics generally considered to be effective antianaerobe drugs were included in the initial regimen did not affect outcome among patients from whom anaerobes were isolated.

Comment by Stan Deresinski, MD, FACP

This was a carefully performed study, the results of which challenge current dogma regarding the role of anaerobes in the etiology of pneumonia in patients with an increased risk of aspiration. It also raises additional questions such as the etiology of the pulmonary infiltrates in the 43% of patients whose cultures were negative. This question was not, however, addressed by El-Solh et al.

Pneumonia occurs via 1 of 3 routes of infection: inhalation, bacteremia, and, by far the most common route, aspiration of oropharyngeal contents. Almost one-fifth of episodes of pneumonia in nursing home residents are reported to be due to aspiration.1 Since most cases of bacterial pneumonia (including, eg, pneumococcal pneumonia) result from aspiration of organisms colonizing the oropharynx, the term "aspiration pneumonia" is, in some ways, redundant and almost meaningless. Nonetheless, this nosologic category has been retained in reference to episodes of pneumonia that occur in patients at increased risk of oropharyngeal aspiration—a circular definition if there ever was one. The problems with considering this a separate category of pneumonia can be further illustrated by the fact that aspiration of oropharyngeal secretions may occur in as many as one-half of healthy adults during sleep.2 Despite these conceptual difficulties, it may be argued that retaining a separate category of pneumonia as due to aspiration is important because aspiration pneumonia implies an important role for anaerobes as a cause of the infection. But even this notion now appears to be bogus!

The putative role of oral anaerobes in aspiration pneumonia was advocated 3 decades ago, largely on the basis of culture of respiratory specimens obtained by transtracheal aspiration.3 As a consequence, it became a standard practice to include an antibiotic with activity against anaerobes in the treatment of "aspiration pneumonia." More recent studies, including the one reviewed here, in which respiratory specimens were obtained from within the lung by use of a protected brush catheter, have failed to demonstrate an important role for anaerobes, despite the use of careful bacteriologic methods.4

The reasons for the discrepancy between the earlier and more recent studies with regard to recovery of anaerobes are probably multiple. For instance, the Bartlett study population was from a Veteran’s hospital and contained a preponderance of alcoholics, their infections were generally more chronic and neglected, and the presence of putrid sputum was commonly reported.4 Pneumonia in nursing home patients is less likely to be long neglected.

It is also possible, perhaps likely, in patients such as the ones in the Bartlett and similar studies, that the tracheal flora being sampled differs from that in the lungs themselves, an observation documented in experiments using dogs.5 These results, in toto, have led to the revisionist statement that "antibiotic agents with specific anaerobic activity are not routinely warranted and may be indicated only in patients with severe periodontal disease, putrid sputum [not verified in the current study], or evidence of necrotizing pneumonia or lung abscess on radiographs of the chest."4


1. Marrie TJ, et al. Nursing home-acquired pneumonia: A case-control study. J Am Geriatr Soc. 1986;34: 697-702.

2. Gleeson K, et al. Quantitative aspiration during sleep in normal subjects. Chest. 1997;111:1266-1127.

3. Bartlett JG, et al. The bacteriology of aspiration pneumonia. Am J Med. 1974;56:202-207.

4. Marik PE. Aspiration pneumonitis and aspiration pneumonia. N Engl J Med. 2001;344:665-671.

5. Moser KM, et al. Sensitivity, specificity, and risk of diagnostic procedures in a canine model of Streptococcus pneumoniae pneumonia. Am Rev Resp Dis. 1982;125:436-442.

Dr. Derenski is Clinical Professor of Medicine, Stanford; Associate Chief of Infectious Diseases, Santa Clara Valley Medical Center.