Abbreviations:CDC Centers for Disease Control and Prevention, FDA Food and Drug Administration, IDSA Infectious Disease Society of America, MIC Minimal inhibitory concentration, VAP ventilator-associated pneumonia
Details concerning relative frequency, pathogenesis, microbiology, diagnosis, and management of anaerobic pleuropulmonary infections were described more than two decades ago [1–6]. In the intervening years antibiotic resistance has become a clinically significant problem, likely as a consequence of the extensive use of antimicrobial agents, which has spurred the development of new ones. However because of the difficulties in both the recovery of obligate anaerobes from clinical specimens and the in-vitro antimicrobial susceptibility testing of anaerobes , the microbial diagnosis frequently remains putative, based on the clinical presentation, especially predisposition of the patient to aspirate or the putrid odor of the breath or infected material (sputum or pleural fluid). Antimicrobial therapy consequently often remains empirical and based on published studies of in-vitro activity against the presumed anaerobic pathogens.
Obligate anaerobic bacteria are organisms that require a low redox potential and oxygen tension for survival and growth. These organisms are the predominant constituents of the normal flora that inhabits the mucosal surfaces of the gingival crevice. The crevicular fluid flora normally is predominantly Gram-positive and the predominant anaerobic species is Actinomyces; but in the presence of periodontitis, the gingival crevice deepens as the epidermal dentine junction recedes toward the tooth apex and the bacterial flora in the abnormal periodontal pocket becomes predominantly Gram-negative . Over 200 microbial species are found in the crevicular fluid in numbers of 1012 CFU/g. The predominant Gram-negative species in the periodontal pocket include Pophyromonas, Prevotella, Bacteroides, and Fusobacterium. The saliva that bathes these mucosal surfaces contains these organisms in addition to other organisms that colonize the nasopharynx, such as obligate anaerobes and known aerobic pulmonary pathogens, such as Streptococcus pneumoniae and Hemophilus influenzae.
Anaerobic pulmonary infection usually is endogenously acquired by aspiration of oropharyngeal contents. Aspiration occurs among normal people, especially during deep sleep, but in certain patients aspiration is thought to be of sufficient magnitude or frequency to overcome lung defenses. The aspirated material may contain adjuvants, such as necrotic tissue, food or foreign bodies, or particularly virulent pathogens, or synergistic combinations of microorganisms. Patients with anaerobic pulmonary infection usually have underlying conditions that predispose to aspiration, such as  loss of consciousness from a seizure, diabetic coma, general anesthesia, head trauma, intoxication due to alcoholism, or other drug abuse or overdose,  cerebrovascular accident, or  esophageal disease.
Anaerobic pulmonary infection developing as a consequence of bronchogenic spread to the lung is usually referred to as primary. Hematogenous anaerobic pulmonary infection is referred to as secondary, because in this instance the primary infection (bacteremia, endocarditis, or suppurutive thrombophlebitis) will usually be clinically evident. Unresolved anaerobic bacterial pneumonitis will undergo necrosis in 1–2 weeks and result in one or more discrete cavities. The necrotizing infection is referred to as a lung abscess if each cavity is 2 cm or more in diameter. If there are multiple, small cavities, each less than 2 cm in diameter, the process is usually referred to as necrotizing pneumonia. Both lung abscess and necrotizing pneumonia are thought to be different manifestations of the same pathogenetic processes. Extension of the anaerobic pulmonary infection to the pleural space may result in anaerobic empyema.
When aspirated into the lower respiratory tract there is a marked simplification of the flora, so that only the most virulent species predominate. About four or five microbial species are isolated from minimally contaminated lower respiratory secretion in these patients. Anaerobic species, such as Prevotella melaninogenica, Fusobacterium nucleatum, and Peptostreptococcus species, usually predominate . Microaerophilic and facultative streptococci are also frequently present. In some patients, the anaerobes may be mixed with facultative respiratory pathogens, such as S. pneumoniae, Staphylococcus aureus, H. influenzae, or Klebsiella pneumoniae. The anaerobic species, unlike facultative microorganisms, typically produce a foul odor in clinical specimens.
Bacteriologic studies in the early 1970s of lower respiratory tract secretions obtained by methods, such as transtracheal aspiration, that minimize contamination with oropharyngeal microbial flora, established that a polymicrobial anaerobic bacterial flora cause the infection in about one-third of patients with communityacquired, one-third of non-intubated patients with nosocomial pneumonia, and in most patients with putrid lung abscess [1–6]. Similarly, pathogens were recovered in 96% of patients with community-acquired pneumonia and anaerobes in 46% of these patients in a study that used fiberoptic bronchoscopy and protected specimen brush to obtain samples from the involved area of lung prior to antimicrobial therapy . Ventilator-associated pneumonia (VAP) is also thought to result from aspiration of oropharyngeal material. Indeed, using rigorous anaerobic methods and protected specimen brush sampling within 24 hours after development of pneumonia, anaerobes were found in 23% of patients with VAP . However, other studies using fiberoptic bronchoscopy and protected specimen brush to obtain cultures have failed to recover anaerobes in aspiration pneumonia and VAP [12,13], perhaps because the patients in these studies received antibiotic therapy prior to bacteriologic sampling .
Anaerobes account for 25–40% of cases in most studies of empyema; the highest frequency of 76% was found in a series of patients with empyema from a large city hospital and two Veterans Administration hospitals .
Patients with primary anaerobic pulmonary infection usually have a predisposition to aspirate and have periodontal disease that would favor the presence of large numbers of potential pathogens and possibly necrotic periodontal tissue in aspirated oropharyngeal secretions. The development of primary anaerobic pulmonary infection in patients who are edentulous should prompt a search for a neoplastic process in the oropharynx or lower respiratory tract that would result in obstruction of the airways and provides a focus of anaerobic microbial proliferation in necrotic neoplastic tissue.
About three-quarters of patients with primary lung abscess have an indolent febrile, wasting illness with respiratory symptoms (cough, sputum production, pleuritic chest pain, blood-streaked sputum) of several weeks duration, similar to that of tuberculosis or lung cancer, except over 50% of patients with lung abscess will produce sputum that has a foul odor or complain of a foul odor to their breath, although in early infection the putrid odor is often absent. One-quarter of patients with lung abscess will present with a more acute illness, similar to that of pneumococcal pneumonia.
The anaerobic etiology of aspiration pneumonia is frequently presumptive, but sputum cultures should be attempted in hospitalized patients to exclude aerobic respiratory pathogens. Febrile patients should have blood cultures. Pleural fluid, if present, should be obtained for stains and cultures. Patients who present with typical features of lung abscess, which include a predisposition for aspiration, periodontal disease, a sputum with foul odor, one or more thick-walled cavities in dependent bronchopulmonary segments (e.g. superior segment of the lower lobe or posterior segment of the upper lobe when aspirating in the supine position) with air-fluid levels, should need little further initial diagnostic work-up and be treated presumptively for a polymicrobial anaerobic infection. In these patients, expectorated sputum is of no value for detecting anaerobes, but may be useful to exclude the presence of other organisms capable of causing pulmonary infection.
Upper lobe cavities without air-fluid levels suggest tuberculosis, and require exclusion of Mycobacterium tuberculosis with three morning sputum collections for mycobacterial stains and cultures. Also in patients who fail to respond to empirical therapy for putative anaerobic pneumonia, in patients suspected of pulmonary neoplasm, and in immunocompromised patients, more rigorous diagnostic testing is usually indicated, including bronchoscopy, protected brushing and collection of bronchoalveolar lavage fluid for stains and cultures for routine bacteria, Rhodococcus equi (in patients with AIDS), legionella, nocardia, and fungi, if expectorated sputum studies fail to disclose the presence of these organisms. Computed tomography of the chest may be important to define pathologic anatomy. Computed tomography, bronchoscopy, bronchoalveolar lavage and lung biopsy may also be necessary in some patients, especially those who do not respond to empirical therapy for lung abscess to exclude non-infectious conditions, such as cystic bronchiectasis, cavitating neoplasms and Wegener’s granulomatosis, which may also produce cavitation and be confused with lung abscess.
Treatment with older antimicrobial agents
Penicillin, either 500–750 mg every 6 hours orally or 10–20 million units intravenously per day, or tetracycline had been the standard antimicrobial agents used in empirical regimens to treat putative anaerobic lung abscess . However, many anaerobic Gram-negative respiratory pathogens are now found to be penicillin-resistant as a consequence of β-lactamase production  and are also tetracycline resistant. Indeed, two studies comparing penicillin to clindamycin in the treatment of putrid lung abscess have found the latter to have a shorter time to defervescence and time to loss of the putrid odor to the sputum and less frequent relapse [15,16]. Clindamycin can be used intravenously in doses of 600 mg every 8 hours initially in hospitalized patients unable to tolerate oral therapy, or orally in doses of 300 mg every 6 hours.
Other older agents that are active against both the oral anaerobes and microacrophilic streptococci include the carbapenems (imipenem and meropenem), cefoxitin, β-lactamase/β-lactam antibiotic combinations (e.g. amoxicillin/clavulanate, ampicillin/sulbactam, ticarcillin/clavulanate or piperacillin/tazobactam), or metronidazole combination [14,17,18]. These drugs, however, do not have the proven efficacy of clindamycin, although susceptibility of anaerobes to clindamycin is no longer as predictable as when this drug was first used . Metronidazole alone has been found to be inadequate therapy for anaerobic pleuropulmonary infections , apparently because it is inactive against microacrophilic streptococci, although it is reliably active against Gram-negative anaerobes . Chloramphenicol is reliably active in vitro against anaerobes [14,17] but, although it has been effective in treatment of anaerobic infections, this drug is rarely used any longer because it can cause fatal aplastic anemia and alternative drugs are available.
Treatment with newer antimicrobial agents
In addition to the older drugs, several new antimicrobial agents have been approved for treatment of respiratory tract infections because of demonstrated efficacy in clinical trials of community-acquired pneumonia. However, none of these drugs has been studied in series of patients with bacteriologically confirmed anaerobic pleuropulmonary infections, or with putative anaerobic infections, such as putrid lung abscess. These drugs have reliable activity against aerobic respiratory pathogens, such as S. pneumoniae and H. influenzae, and Moraxella catarrhalis, which have emerged resistant to the older agents. They are also active against the so-called ‘atypical’ pathogens, Chlamydia pneumoniae, Mycoplasma pneumoniae and Legionella pneumophila. The in-vitro activity demonstrated against a variety of anaerobic bacteria suggests that these drugs may have clinical efficacy in anaerobic pleuropulmonary infections. However, the sources for anaerobic isolates in these studies have not necessarily been lower respiratory tract secretions or pleural fluid; when the source is other than pleuropulmonary, the published data may not be predictive of efficacy for infection caused by anaerobic pleuropulmonary pathogens.
The older fluoroquinolones, such as ciprofloxacin, ofloxacin, norfloxacin, enoxacin, and lomefloxacin, are inactive against anaerobes. Trovafloxacin, gatifloxacin, moxifloxacin and gemifloxacin, and to a limited extent sparfloxacin and levofloxacin, are active against anaerobes [20,21,22•,23,24•,25]. Use of trovafloxacin is severely limited by the rare occurrence of hepatotoxicity that has been fatal in some cases, and use of sparfioxacin is limited by the occurrence of phototoxicity. Side effects with levofloxacin, gatifloxacin and moxifloxacin have been minimal, although prolongation of the Q–Tc interval in some patients with sparfioxacin, gatifloxacin, and moxifloxacin has resulted in the recommendation that their use be avoided in patients with known prolongation of the Q–Tc interval, patients with uncorrected hypokalemia, and patients receiving class 1A or class III antiarrhythmic agents. Food and Drug Administration (FDA) approval of gemifloxacin is still pending.
Because of their broad spectrum of activity for respiratory pathogens and infrequency of significant side effects, these newer fluoroquinolones have been used commonly for empirical therapy of pneumonia. Indeed the latest Infectious Disease Society of America (IDSA) guidelines for the treatment of community-acquired pneumonia [26••] recommended their use, either (1) alone for treatment of outpatients or (2) combined with a β-lactam, such as ceftriaxone, for the treatment of patients requiring intensive care unit admission. Combination therapy was recommended because their efficacy when used alone in severe pneumonia has not been documented [26••]. For suspected aspiration, the IDSA recommended a newer fluoroquinolone, with or without a β-lactam/β-lactamase inhibitor combination, such as ampicillin/sulbactam or piperacillin/tazobactam, metronidazole, or clindamycin. They specifically commented, though, that the newer fluoroquinolones might not require additional anaerobic coverage, because of their in-vitro activity against anaerobes.
A Centers for Disease Control and Prevention (CDC) panel that was recently convened to consider treatment of community-acquired pneumonia in the face of emerging antimicrobial resistance was hesitant to recommend the newer fluoroquinolones for fear that their widespread use may lead to development of fluoroquinolone resistance among the respiratory pathogens (as well as other pathogens colonizing the treated patients) [27••]. The CDC panel did not specifically comment on treatment of anaerobic pulmonary infection.
The ketolides are a new family of antimicrobial agents that are structurally related to the macrolides. Their mechanism of action, similar to that of the macrolides, involves blocking protein synthesis by binding to the ribosomal RNA complex; but bacteria that either display MLSb type of resistance (i.e. cross-resistant to macrolides, clindamycin and streptogramin B) as a result of decreased affinity to an altered ribosomal target or display efflux type of resistance (i.e. resistant to macrolides, but susceptible to clindamycin) are still susceptible to the ketolides. Telithromycin, a ketolide whose approval by the FDA for therapy of respiratory tract infection is pending, has been shown to be active against streptococci and staphylococci resistant to the macrolides. This ketolide is also active against many anaerobes, with minimal inhibitory concentration (MIC) ≤ 0.5 μg/ml [28•]. Telithromycin is well absorbed and, when given in doses of 800 mg every 24 hours p.o., achieves peak and trough serum levels of about 2 μg/ml and 0.05 μg/ml, respectively, and has a serum half-life of 10 hours. Iris concentrated ≥ 100-fold within phagocytes and about 5-fold in the epithelial lining fluid of the alveoli.
Linezolid, a member of a new class of synthetic antimicrobial agents, the oxazolidinones, has been approved by the FDA for therapy of respiratory tract infection. Because of its unique mechanism of action, cross-resistance with other antimicrobial agents does not occur. Linezolid has been found to be active against Fusobacterium spp., Prevotella spp., Porphyromonas spp., Bacteroides spp., and peptostreptococci at MIC ≤ 2 μg/ml [29•]. Linezolid is given in doses of 600 mg i.v. or p.o. every 12 hours. Its peak and trough serum levels are 12 and 4 μg/ml, respectively, and its serum half-life is about 4 hours. Doses do not have to be changed for renal failure. Side effects have been minimal, although reversible thrombocytopenia has been reported in about 3% of treated patients.
Patients on effective therapy will defervesce and in patients with putrid lung abscess the sputum will become odorless within a week. Chest radiographic findings may take several weeks to resolve. The duration of therapy is controversial. The recommended duration of therapy for aspiration pneumonia is 14 days and for anaerobic lung abscess it is usually at least 4–6 weeks to prevent relapse, or until the abscess completely resolves or there is a small, stable residual scar. However, many patients will cease to take medication on their own after several weeks and return at a later date for other reasons with a clear chest radiogram. Failure to adequately respond to therapy (persistent fever, worsening of the pulmonary radiographic findings, or development of empyema) should prompt more intensive investigation to exclude the presence of resistant pathogens or a non-infectious etiology.
Surgery is rarely indicated for putrid lung abscess, except for the rare complication of massive hemoptysis. Postural and bronchoscopic drainage and chest physiotherapy traditionally have been recommended to enhance antimicrobial therapy. However caution should be exercised to avoid sudden massive emptying of pus-filled cavities into the airways and previously uninvolved bronchopulmonary segments. Anaerobic empyema will require complete drainage usually by insertion of a chest tube, although thoracotomy and rib resection may be necessary to break loculations to accomplish this.
Obligate anaerobes are the predominant constituents of normal oropharyngeal flora and produce pleuropulmonary infection in patients who are prone to aspirate. The diagnosis and therapy of anaerobic pulmonary infection is frequently empirical and guided by published studies of in-vitro activity against collected clinical isolates. Several new drugs with in-vitro activity against obligate anaerobes have recently become available for empirical treatment of pneumonia.
References and recommended reading
Papers of particular interest, published within the annual period of review, have been highlighted as:
• of special interest
•• of outstanding interest
1. Levison ME, Bran JL, Ries K. Treatment of anaerobic bacterial infections with clindamycin 2 phosphate. Antimicrob Agents Chemother 1974; 5:276–280.
2. Cesar L, Gonzalez C, Calia FM. Bacteriologic flora of aspiration-induced pulmonary infections. Arch Intern Med 1975; 135:711–714.
3. Lorber B, Swenson RM. Bacteriology of aspiration pneumonia, a prospective study of community and hospital acquired cases. Ann Intern Med 1974; 81:329.
4. Bartlett JG, O’Keefe P, Tally FP, et al.
Bacteriology of hospital-acquired pneumonia. Arch Intern Med 1986; 146: 868–871.
5. Bartlett JG, Gorbach SL, Finegold SM. The bacteriology of aspiration pneumonia. Am J Med 1974; 56:202–207.
6. Bartlett JG, Gorbach SL. Treatment of aspiration pneumonia and primary lung abscess: penicillin G vs. clindamycin. JAMA 1975; 234:935–937.
7. Jorgensen JH, Ferraro MJ. Antimicrobial susceptibility testing: special needs for fastidious organisms and difficult to detect resistance mechanisms. Clin Infect Dis 2000; 30:799–808.
8. Slots J. Subgingival microflora and periodontitis. J Clin Periodont 1979; 6:351–382.
9.•• Marina M, Strong CA, Civen R, et al.
Bacteriology of anaerobic pleuropulmonary infection. Preliminary Report. Clin Infect Dis 1993; 16 (suppl 4):S256–S262. Susceptibility of a large panel of anaerobes that have caused pleuropulmonary infection to a variety of older antimicrobial agents.
10. Pollack HM, Hawkins EL, Bonner JR, et al.
Diagnosis of bacterial pulmonary infections with quantitative protected catheter cultures obtained during bronchoscopy. J Clin Microbiol 1983; 17:255–259.
11. Dore P, Robert R, Grolier G, et al.
Incidence of anaerobes in ventilator-associated pneumonia with use of a protected specimen brush. Am J Resp Crit Care Med 1996; 153:1292–1298.
12. Mier L, Dreyfuss D, Darchy B, et al.
Is penicillin G adequate initial treatment for aspiration pneumonia? A prospective evaluation using a protected specimen brush and quantitative cultures. Intensive Care Med 1993; 19:279–284.
13. Marik PE, Careau P. The role of anaerobes in patients with ventilator-associated pneumonia and aspiration pneumonia, a prospective study. Chest 1999; 115:178–183.
14. Bartlett JG. Anaerobic bacterial infections of the lung and pleural space. Clin Infect Dis 1993; 16 (suppl 4):S248–S255.
15. Levison ME, Mangura CT, Lorber B, et al.
Clindamycin compared with penicillin for the treatment of anaerobic hung abscess. Ann Intern Med 1983; 98:466–471.
16. Gudiol F, Manresa F, Pallares R, et al.
Clindamycin vs. penicillin for anaerobic hung infections. High rate of penicillin failures associated with penicillin-resistant Bacteroides melaninogenicus.
Arch Intern Med 1990; 150:2525–2529.
17. Kasten MJ. Clindamycin, metronidazole and chloramphenicol. Mayo Chin Proc 1999; 74:825–833.
18. Falagas ME, Siakavellas E. Bacteroides, Prevotella
species: a review of antibiotic resistance and therapeutic options. Int J Antimicrob Agents 2000; 15:1–9.
19. Perlino CA. Metronidazole vs. clindamycin treatment of anaerobic pulmonary infection: failure of metronidazole therapy. Arch Intern Med 1981; 141:1424–1427.
20. Appelbaum PC. Quinolone activity against anaerobes. Drugs 1999; 58:60–64.
21. Schaumann R, Ackerman G, Pless B, et al.
In-vitro activities of gatifoxacin, two other quinolones, and five nonquinolone antimicrobials against obligately anaerobic bacteria. Antimicrob Agents Chemother 1999; 43:2783–2786.
22.• Goldstein EJ. Review of the in vitro activity of gemifloxacin against Gram-positive and Gram-negative anaerobic pathogens. J Antimicrob Chemother 2000; 45:55–65. Susceptibility of a panel of anaerobes to gemifloxacin.
23. Goldstein EJ, Citron DM, Vreni Merriam C, et al.
Activities of gemifloxacin (SB 265805, LB 20304) compared to those of other oral antimicrobial agents against unusual anaerobes. Antimicrob Agents Chemother 1999; 43:2726–2730.
24.• Goldstein EJ, Citron DM, Merriam CV, et al.
Activity of gatifloxacin compared to five other quinolones versus aerobic and anaerobic isolates from skin and soft tissue samples of human and animal bite wound infections. Antimicrob Agents Chemother 1999; 43:1475–1479. Susceptibility to newer fluoroquinolones of a panel of anaerobes that have caused bite wound infection.
25. Ackerman G, Schaumann R, Pless B, et al.
Comparative activity of moxifloxacin in vitro
against obligately anaerobic bacteria. Eur J Clin Microbiol Infect Dis 2000; 19:228–232.
26.•• Bartlett JG, Dowell SF, Mandell LA, et al.
Practice guidelines for the management of community-acquired pneumonia in adults. Clin Infect Dis 2000; 31:347. Infectious Disease Society of America’s state of the art review of diagnosis and management of community-acquired pneumonia.
27.•• Heffelfinger JD, Dowell SF, Jorgensen JH, et al.
Management of community-acquired pneumonia in the era of pneumococcal resistance. A report from the drug-resistant Streptococcus pneumoniae
therapeutic working group. Arch Intern Med 2000; 160:1400. CDC expert panel’s thoughtful analysis of options for empirical treatment of community-acquired pneumonia in the face of multi-drug resistant pneumococci.
28. Goldstein EJ, Citron DM, Merriam CV, et al.
Activities of telithromycin (HMR 3647, RU 66647) compared to those of erythromycin, azithromycin, clarithromycin, roxithromycin, and other antimicrobial agents against unusual anaerobes. Antimicrob Agents Chemother 1999; 43:2801–2805. Ketolide susceptibility of a panel of anaerobes from a variety of sources.
29.• Goldstein F, Citron DM, Merriam CV. Linezolid activity compared to those of other selected macrolides and other agents against aerobic and anaerobic pathogens isolated from soft tissues bite infections in humans. Antimicrob Agents Chemother 1999; 43:1469–1474. Linezolid susceptibility of a panel of anaerobes that have caused bite wound infection.