Anaerobic bacterial infections

INTRODUCTION — Anaerobic bacteria are the predominant cultivable and noncultivable bacteria in the human body and can be recovered from infections at virtually all anatomic sites, although the frequency of recovery is highly variable. This topic will focus on the microbiology, pathogenesis, sites of infection, diagnosis, and management of obligate anaerobic flora that are endogenous to the host.

Actinomycosis is discussed separately (see "Cervicofacial actinomycosis" and "Abdominal actinomycosis" and "Treatment of actinomycosis"). Toxin-mediated and/or tissue destructive clostridial infections are reviewed separately by individual pathogen and/or syndrome (eg, tetanus, botulism, Clostridioides [formerly Clostridium] difficile infection, gas gangrene, neutropenic enterocolitis due to Clostridium septicum). (See "Tetanus" and "Botulism" and "Clostridioides difficile infection in adults: Epidemiology, microbiology, and pathophysiology" and "Clostridial myonecrosis" and "Toxic shock syndrome due to Paeniclostridium sordellii" and "Neutropenic enterocolitis (typhlitis)".)

Fusobacteria are reviewed separately by syndrome (eg, Lemierre syndrome, Ludwig angina, deep neck space infections, liver abscess). (See "Lemierre syndrome: Septic thrombophlebitis of the internal jugular vein" and "Deep neck space infections in adults" and "Epidemiology, pathogenesis, and clinical manifestations of odontogenic infections" and "Ludwig angina" and "Pyogenic liver abscess".)

Discussion of microbiome and the host-pathogen relationship in the gastrointestinal tract is found elsewhere. (See "Spatial organization of intestinal microbiota in health and disease".)

MICROBIOLOGY

Definition — Anaerobic bacteria are defined as bacteria that require reduced oxygen tension for growth. These bacteria fail to show surface (colony) growth in 10 percent carbon dioxide (eg, a microaerophilic environment that supports Campylobacter spp growth) or in air (18 percent oxygen) [1].

Classification by aerotolerance — Anaerobes may be classified by their relative aerotolerance.

●Obligate (strict) anaerobes − Obligate anaerobes are rapidly killed by oxygen and grow in the complete absence of oxygen. The anaerobic genera Bacteroides spp, Fusobacterium spp, Prevotella spp, gram-positive anaerobic cocci, and most Clostridia spp are considered obligately anaerobic. Even strict anaerobes may survive, but usually not grow, in small amounts of oxygen (eg, nanomolar levels) as seen with B. fragilis [2].

●Aerotolerant anaerobes − Aerotolerance among anaerobes varies by species with some species tolerating up to 8 percent oxygen [1]. While some of these organisms survive in air, they do not replicate and are metabolically inactive in this state. Examples of clinically important aerotolerant anaerobes are some Clostridia spp, and most Cutibacterium spp (eg, including Cutibacterium acnes).

Identification — Anaerobic bacteria are fastidious organisms that can be difficult to grow in clinical microbiology laboratories. These organisms require low-oxygen conditions for growth and most grow slowly in complex communities. In these communities, pathogenic anaerobic bacteria most often coexist with commensal or symbiotic anaerobic bacteria, further hindering identification of potential anaerobic pathogens from complex specimens (eg, stool, abscesses). (See 'Microbiologic diagnosis' below.)

While most clinical laboratories can identify the major clinically significant anaerobes (eg, Clostridium spp, Bacteroides spp, Fusobacterium spp, Peptostreptococcus spp, Cutibacterium spp), growing and identifying the majority of colonizing anaerobic bacteria is challenging using laboratory methods. As an example, in an analysis of 13,555 prokaryotic ribosomal RNA gene sequences from the colon, most bacteria were considered novel, uncultivated micro-organisms [3].

Clinically significant anaerobes — The major clinically significant anaerobes are summarized in the table (table 1). Most clinical laboratories should be able to identify these organisms. Clinically significant anaerobes can be divided into gram-negative bacilli, spore-forming and non-spore-forming gram-positive bacilli, and gram-positive or gram-negative cocci.

●Gram-negative bacilli – Important gram-negative anaerobes for identification include the Bacteroides spp, Prevotella spp, and Fusobacterium spp.

•Bacteroides spp – Bacteroides spp are small, pleomorphic gram-negative bacilli. Among the more than 50 known Bacteroides spp [4], B. fragilis is the most important in human disease. Other Bacteroides spp, previously designated with B. fragilis as the 'B. fragilis group' (ie, B. thetaiotaomicron, B. distasonis, B. ovatus, and B. vulgatus), also contribute to human disease. Bacteroides spp are distinguished by their ability to grow in the presence of 20 percent bile.

B. fragilis is a uniquely virulent anaerobe. The organism typically constitutes only 0.5 percent or less of cultivable normal colonic flora, yet it is the most common anaerobe isolated from intra-abdominal infections and is the most frequent anaerobic isolate in cases of bacteremia [5-7].

•Prevotella spp – Pigmented Prevotella (eg, P. melaninogenica, P. corporis, P. intermedia, P. loescheii, and P. nigrescens) are a clinically important group of fastidious anaerobes. The frequency of recovery of these species depends upon the expertise of the laboratory. Other pigmented anaerobic bacteria include species from the genera Porphyromonas spp and Alistipes spp.

•Fusobacterium spp – Fusobacteria are fastidious organisms that can be distinguished from Bacteroides spp because they are usually indole positive and nonmotile and show distinctive morphologic features on Gram stain (typically slender, elongated gram-negative rods but can be gram-variable as well as variable in morphology).

●Spore-forming gram-positive bacilli – Spore-forming gram-positive bacilli includes Clostridium spp and Clostridioides difficile. The species are heterogenous and are known for their toxin-producing properties and difficult to eradicate spores that are visible on Gram stains. Clostridia spores enable the organism to survive exposure to ethanol for 30 minutes or heating to 80°C for 10 minutes. Extensive biochemical testing is required for speciation, which generally is unnecessary except in selected cases, such as gas gangrene. C. difficile may be cultured from stool, but antigen assays, toxin assays, and polymerase chain reaction are more practical.

●Non-spore-forming gram-positive bacilli – The non-spore-forming gram-positive bacilli group includes genera from the phyla Actinobacteria and Firmicutes. Actinobacteria genera of clinical relevance include Actinomyces spp, Cutibacterium spp (formerly Propionibacterium), and Eggerthella spp [8]; Firmicutes genera include Faecalibacterium spp and Lactobacillus spp among others. This complex group of bacteria is most often identified by DNA sequencing analysis and/or matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS). With the exception of Actinomyces spp, most of these organisms have limited documented pathogenic potential. While Cutibacterium spp is usually a contaminant in clinical samples, this bacterium can play an important role in septic arthritis complicating shoulder prostheses, infections following neurosurgery, and other device-related infections [9]. Actinomyces is discussed separately. (See "Cervicofacial actinomycosis" and "Abdominal actinomycosis" and "Treatment of actinomycosis".)

●Gram-positive cocci – The key clinically relevant anaerobic gram-positive cocci are Peptostreptococcus spp. The species rarely merits speciation because these organisms lack distinctive virulence properties and appear to be predictably susceptible to antimicrobial agents.

●Gram-negative cocci – Veillonella species (eg, Veillonella parvula) are common anaerobic members of the oral flora.

Susceptibility testing — Anaerobic culture results with susceptibility testing are often labor intensive, time consuming, expensive, incomplete, and generally not available in a clinically relevant time frame. Thus, antibiotic decisions for anaerobic pathogens are often best made based upon predicted susceptibility patterns according to the Gram stain, the anatomic site of infection, and available published survey data for specific anaerobes [10-12].

Based on these practical issues, the Clinical and Laboratory Standards Institute (CLSI) Working Group on Anaerobic Susceptibility Testing and the Manual of Clinical Microbiology provide guidance on anaerobe susceptibility testing [12,13]. The following settings are clinical situations in which anaerobic susceptibility testing may be reasonable for anaerobes considered pathogenic.

●To monitor local and regional resistance patterns

●To test new antimicrobial agents

●To select antibiotics that are critical for individual patient management because of:

•Critical illness

•Known resistance of an organism or species

•Persistent infection despite appropriate antibiotics

•Anticipated pivotal role of the selected antimicrobial agent in therapy (eg, osteomyelitis, joint space infections)

•To confirm antimicrobial activity when long courses of antibiotics are required

Most clinical laboratories will not perform anaerobe susceptibility tests unless they are specifically requested. Because anaerobes are slow growing, results are usually available only after therapeutic decisions have already been made.

ROLE IN THE HUMAN MICROBIOME — Anaerobic bacteria are major constituents of the human microbiome at virtually all anatomic sites. Viruses, fungi, and parasites are also microbiome members although their contributions to human health and microbiome function remain less well studied and understood. Most mucocutaneous surfaces of humans harbor a rich indigenous flora composed of aerobic and anaerobic bacteria, with microbial species and concentrations varying at different anatomic sites.

The assembly of the human microbiome begins at birth and is impacted by mode of birth. It appears that even these very early facets of microbiome assembly are linked to disease onset in children [14]. Most studies have explored the makeup of human colonic bacterial microbiota. The colon accounts for the greatest abundance of human anaerobic bacteria (500 to 1000 species per individual) [3]. The colon microbiome functions to maintain health by establishing ecologic balance, assisting with nutrition, and preventing colonization with exogenous organisms including potential human disease pathogens ('colonization resistance'). There is substantial individual variation that is strongly influenced by diet and medications (especially antibiotics) with less impact by host genetic make-up [3,15,16]. (See "Spatial organization of intestinal microbiota in health and disease".)

Any oral antibiotic exposure disrupts the colon microbiome and longer term effects of repeated antibiotic exposure on disease are under study [17,18]. For example, oral antibiotics decrease the 'colonization resistance' provided by a diverse colon microbiota, which enhances the potential for infection with enteric pathogens and colonization by resistant bacteria, such as vancomycin-resistant enterococci, and resistant gram-negative bacilli. These bacteria cause many nosocomial infections, including serious infections in patients with neutropenia, organ transplantation, or critical illness. Disruption of the colon microbiome is critical to precipitating C. difficile infection and linked, for example, to outcomes in bone marrow transplantation [19]. (See "Overview of neutropenic fever syndromes" and "Nosocomial infections in the intensive care unit: Epidemiology and prevention" and "Clostridioides difficile infection in adults: Clinical manifestations and diagnosis".)

While extensive work seeks to identify approaches to modify the microbiome to impact either acute or chronic disease, only limited approaches are available for clinical use, namely, fecal microbiota transplantation (FMT), most often used for treatment of recurrent C. difficile colitis, as well as FDA-approved fecal microbiota products for prevention of recurrent C. difficile colitis [20,21]. (See "Overview of possible risk factors for cardiovascular disease", section on 'Trimethylamine-N-oxide' and "Probiotics for gastrointestinal diseases", section on 'Irritable bowel syndrome' and "Fecal microbiota transplantation for treatment of Clostridioides difficile infection".)

PATHOGENESIS — Anaerobic infections nearly always arise from spillage and/or invasion of endogenous bacteria into contiguous sites of the body but also spread systemically (eg, bacteremia). The usual pathophysiologic mechanism for anaerobic infection is a breach in the mucocutaneous barrier (eg, bowel perforation, carcinoma with obstruction, mucositis, perirectal lesions, or compromised consciousness with aspiration) resulting in displacement of the resident flora.

Important exceptions are toxin-mediated syndromes such as clostridial syndromes, including botulism, Clostridium perfringens food poisoning, enteritis necroticans, tetanus, some cases of gas gangrene, and Clostridioides difficile-associated diarrhea.

The following are major factors implicated in the pathogenesis of anaerobic infections:

●Production of toxins — The most clearly identified virulence factors for anaerobic bacteria are the exotoxins produced by clostridial species, including botulinum toxins, tetanus toxin, C. difficile toxin B, and the >20 toxins produced by C. perfringens (as well as many other clostridial species) [22-24]. Clostridial species have also been shown to produce leukocytosis-inducing factors that result in marked leukocytosis in some infected patients ("leukemoid reaction") [25].

A molecular subgroup of Bacteroides fragilis secrete a pro-inflammatory metalloprotease enterotoxin [26]. Strains producing the toxin are associated with self-limited diarrheal illnesses in children and adults and not uncommonly asymptomatically colonize healthy individuals. There is evidence from mouse models and human studies that enterotoxin-producing strains of B. fragilis may play a role in colon carcinoma [27-29].

●Role of capsular polysaccharides in abscess formation — Capsular polysaccharide complexes have zwitterionic properties (presenting oppositely charged atoms) that are important for abscess formation [30,31].

●Presence of lipopolysaccharides — Anaerobic gram-negative bacteria, like all gram-negative bacteria, contain lipopolysaccharide (LPS) that can be extracted from the cell envelope. LPS from anaerobic bacteria can differ structurally from LPS from aerobic gram-negative bacteria and may display lower biologic activity [32]. The endotoxin of Fusobacteria is postulated to account, at least in part, for the severity of illness associated with Lemierre disease [33,34].

●Ability to tolerate oxygen — Some anaerobic bacteria, such as the virulent B. fragilis, can tolerate exposure to oxygen and even replicate at very low oxygen tensions [35]. The ability to survive exposure to oxygen facilitates the survival and thus pathogenicity of the organism such as in abdominal abscess formation. Furthermore, aerobic bacteria are thought to facilitate the growth of B. fragilis and other anaerobes by utilizing the oxygen in the environment [36].

●Production of short chain fatty acids — Anaerobes produce short-chain fatty acids that are thought to contribute to intestinal homeostasis with additional systemic effects, often through G-protein-coupled receptors. [37,38].

SITES OF INFECTION

Spectrum of disease — Anaerobes are part of the normal skin, genital, oral, and gastrointestinal tract (especially the colon) flora. Infections often originate from these sites and contiguously spread. Infections range from relatively minor skin and oral infections (eg, pharyngitis and gingivitis) to life-threatening invasive infections (eg, bacteremia, distant focal suppurative metastatic infections involving the brain, deep neck space, or muscle and fascia) [39-42] (table 2).

In many cases, the infections are polymicrobial, involving other microbes present at the tissue site.

Presence of an abscess should alert the clinician to the likely presence of anaerobes, as anaerobic infections are frequently associated with abscess formation.

Clinical manifestations, diagnosis, and treatment for given infectious syndromes are discussed in relevant UpToDate topics, for which links are provided below.

Infections of the upper airways — Anaerobic bacteria comprise 90 percent of the oral cavity bacteria [1] and are involved in a variety of infections of the oral cavity and adjacent structures. Organisms isolated in oral infection reflect the normal oral flora. The dominant isolates are Fusobacterium spp, Bacteroides spp, gram-positive anaerobic cocci (eg, Peptostreptococci), Veillonella spp, the pigmented Prevotella spp, Porphyromonas spp, Treponema spp, microaerophilic streptococci, and aerobic streptococci [43-45] (table 2).

Dental infections — Anaerobes are usually involved in gingivitis, periodontitis, pyorrhea, pulpitis (endodontal infection), periapical or dental abscess, and perimandibular space infections (table 2). These three latter infections usually represent a continuum: the initial lesion is endodontal, then the infection progresses to the periapical region, and then may extend through the mandible to involve the potential spaces created by fascial insertions along the mandible [46-48]. Odontogenic sinusitis with orbital and intracranial extension is a complication of apical periodontitis [49]. (See "Epidemiology, pathogenesis, and clinical manifestations of odontogenic infections".)

An infrequent but distinct oral infection is necrotizing ulcerative gingivitis, sometimes known as Vincent angina or trench mouth [47]. This is a fulminant infection associated with severe pain, tissue destruction, pseudomembrane formation, and putrid discharge. The bacterial agent is not well established but Fusobacteria spp may contribute and anaerobic spirochetes (eg, Treponema spp) have been detected within tissue at the advancing edge of inflammation. Antibiotic treatment directed against anaerobes is necessary. (See "Epidemiology, pathogenesis, and clinical manifestations of odontogenic infections" and "Overview of gingivitis and periodontitis in adults".)

A possibly related necrotizing infection of the oral mucous membranes is noma, also called cancrum oris or orofacial gangrene, which is characterized by destruction of soft tissue and bone. It evolves rapidly from gingival inflammation to orofacial gangrene [50]. Noma occurs most frequently in children (peak incidence at ages 1 to 4 years) with malnutrition or systemic disease, is usually fatal in the absence of antibiotic therapy, and is considered a neglected disease [51]. The microbiology is poorly documented but anaerobic gram-negative rods including Fusobacteria spp (eg, F. necrophorum and others) are thought to contribute. (See "Noma (cancrum oris)".)

Deep neck space infections — Deep neck space infections usually arise from dental infections involving molar teeth and, less commonly, from infections of the pharynx or tonsils. Oral flora (including oral anaerobes) play an important role in these infections, but they may also be caused by Staphylococcus aureus and aerobic enteric gram-negative bacilli [52] (table 2).

Clinically important anaerobic infections of the deep neck space include:

●Peritonsillar abscess – Peritonsillar abscess is an example of a deep neck space infection that is frequently caused by anaerobic bacteria, such as F. necrophorum and Peptostreptococcus spp. (See "Peritonsillar cellulitis and abscess".)

●Ludwig angina – Ludwig angina is an infection characterized by bilateral involvement of the sublingual, submandibular, and submental spaces that causes swelling of the base of the tongue and potential airway compromise, hence "angina." It is typically a polymicrobial infection involving the flora of the oral cavity, including anaerobes [53]. (See "Ludwig angina".)

●Lemierre syndrome – Lemierre syndrome, or jugular vein suppurative thrombophlebitis, is an infection involving the posterior compartment of the lateral pharyngeal space complicated by suppurative thrombophlebitis of the jugular vein. This life-threatening infection is most often associated with Fusobacterium necrophorum bacteremia, with potential for metastatic abscesses, primarily to the lung [54]. Other anaerobic species (eg, Fusobacterium nucleatum, streptococci) occasionally are isolated [55]. (See "Lemierre syndrome: Septic thrombophlebitis of the internal jugular vein".)

Studies have also implicated F. necrophorum as a possibly important cause of pharyngitis [56-58]. Deep neck space infections are discussed in greater detail separately. (See "Deep neck space infections in adults".)

Chronic rhinosinusitis — Anaerobic bacteria have been implicated with variable frequencies in chronic sinusitis, chronic otitis media, and mastoiditis but play a minimal role in acute otitis media or acute sinusitis (table 2). (See "Microbiology and antibiotic management of chronic rhinosinusitis" and "Chronic rhinosinusitis: Clinical manifestations, pathophysiology, and diagnosis".)

Pleuropulmonary infections — Anaerobic bacteria (mostly from the oral anaerobic flora) are relatively common and frequently overlooked pathogens in the lower airways [44] (table 2). These infections usually arise from aspiration of oral secretions that result in aspiration pneumonitis. Classically, this is an indolent form of pneumonia, but the early presentation is difficult to distinguish from other forms of acute bacterial pneumonia since the clinical clues indicating an anaerobic infection are typically absent early on. Pathogens in macroaspiration pneumonia depend on the clinical context and can include anaerobes (predominantly oral anaerobes), aerobic bacteria, and nosocomial pathogens [59,60].

Although now seen infrequently, clinical clues to chronic anaerobic aspiration pneumonia include a predisposition to aspirate, infection in a dependent pulmonary segment, putrid sputum, and an indolent course. Whereas patients with pneumococcal pneumonia often have the abrupt onset of symptoms accompanied by a shaking chill and a rapid progression of symptoms, these features are rare with an anaerobic pleuropulmonary infection. By contrast, patients with a chronic aspiration pneumonia and/or pulmonary abscess often present with weight loss, anemia, and chronic pulmonary complaints, all features that are relatively uncommon in pneumonia due to most aerobic bacteria other than mycobacteria. Nevertheless, pneumonia due to anaerobic bacteria may simulate pneumococcal pneumonia during the early stages and the clinical clues that lead clinicians to suspect anaerobic infection can be absent [44,59]. (See "Aspiration pneumonia in adults" and "Lung abscess in adults", section on 'Microbiology' and "Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults", section on 'Bacteria'.)

Intra-abdominal infections — Anaerobic infections within the abdominal cavity are usually polymicrobial and result in intra-abdominal abscess or secondary peritonitis, which may be generalized or localized (phlegmon).

Organisms isolated in intra-abdominal infections typically reflect normal colon flora. The stomach is protected by the gastric acid barrier and consequently harbors relatively small numbers of bacteria. The small bowel also has a limited microbial community, in part, likely due to its rapid motility and higher oxygen content that limit stable anaerobic bacterial colonization. In contrast, if this flow is disrupted (eg, stagnant small bowel segment due to stricture, obstruction, diverticulum, blind loop), high concentrations of bacteria result with a predominance of anaerobes [61]. This bacterial overgrowth pattern may cause malabsorption (see "Small intestinal bacterial overgrowth: Etiology and pathogenesis" and "Pathophysiology of irritable bowel syndrome", section on 'Alteration in fecal microflora'). The largest concentrations of anaerobic bacteria are found in the relatively stagnant terminal ileum and colon, where concentrations reach 10¹¹ per gram, and anaerobic bacteria account for approximately 99.9 percent of the bacteria present [3,62,63]. The most important and frequent anaerobic bacteria are Bacteroides spp, Prevotella spp, Clostridium spp, and Peptostreptococcus spp (table 2).

Intra-abdominal infections are discussed in detail separately. (See "Antimicrobial approach to intra-abdominal infections in adults" and "Acute appendicitis in children: Clinical manifestations and diagnosis" and "Acute appendicitis in adults: Clinical manifestations and differential diagnosis".)

Infections of the female genital tract — Nearly all infections of the female genital tract that are not caused by sexually transmitted pathogens are likely to involve anaerobic bacteria (table 2). This includes pyometra (accumulation of pus in the uterine cavity), tubo-ovarian abscess, amnionitis, septic abortion or gynecologic procedures, endometritis, septic pelvic thrombophlebitis, and some wound infections after gynecologic surgery. Pelvic inflammatory disease (PID) is most often an ascending infection caused by anaerobic and facultative aerobic vaginal bacteria including those causing bacterial vaginosis; less than half of PID cases are associated with sexually transmitted pathogens (eg, Neisseria gonorrhoeae, Chlamydia trachomatis). PID predisposes to infertility [64-67].

The most common anaerobes are aerobic, microaerobic, and anaerobic lactobacilli, Peptostreptococcus spp, Bacteroides spp, and penicillin-resistant anaerobes such as Prevotella bivia and Prevotella disiens (formerly Bacteroides bivius and Bacteroides disiens). There are emerging associations of Fusobacterium nucleatum, Porphyromonas gingivalis, and Sneathia infections with pregnancy and adverse outcomes of pregnancy [68-70]. Factors that appear to influence the bacteriologic findings in the genital tract include menarche, menopause, pregnancy, antibiotic therapy, new sexual partners, male circumcision, hormonal therapy, black race, douching, smoking, and lubricant use [71].

The contribution of the genital tract microbiota to homeostasis and disease (eg, susceptibility to sexually transmitted pathogens) and approaches to modifying the genital tract microbiota are under study [71]. Bacterial vaginosis (BV) is a complex dysbiosis of the genital tract flora, with a decrease in lactobacilli concentrations and a predominance of anaerobes [72].

Female genital tract infections are discussed in more detail elsewhere. (See "Bacterial vaginosis: Clinical manifestations and diagnosis", section on 'Pathogenesis and microbiology' and "Vulvar abscess", section on 'Microbiology' and "Pelvic inflammatory disease: Pathogenesis, microbiology, and risk factors" and "Epidemiology, clinical manifestations, and diagnosis of tubo-ovarian abscess", section on 'Microbiology' and "Clinical chorioamnionitis", section on 'Microbiology' and "Septic abortion: Clinical presentation and management", section on 'Epidemiology and microbiology' and "Postpartum endometritis", section on 'Microbiology' and "Septic pelvic thrombophlebitis", section on 'Microbiology' and "Complications of gynecologic surgery", section on 'Infectious morbidity'.)

Soft tissue infections — Anaerobic bacteria are common pathogens in a diverse array of skin and soft tissue infections, such as infected sebaceous (epidermal) cysts, infected pilonidal cysts, paronychia, wound infections after surgery, bite wounds, diabetic foot ulcers, and decubitus ulcers [39,73-79]. While Staphylococcus aureus and Streptococcus pyogenes are commonly viewed as the dominant pathogens in soft tissue infections, anaerobic bacteria account for a major portion of some cutaneous abscesses (eg, pilonidal cysts, bite wounds, diabetic foot infections) and may be playing an important role in hidradenitis suppurativa [80].

The skin flora contains large numbers of anaerobic bacteria. The predominant organisms are Cutibacterium (formerly Propionibacterium) acnes and, to a lesser extent, other species of Cutibacterium spp and Peptostreptococcus spp [81]. Most infections involve the cutaneous or adjacent mucosal surface flora (table 2). Cutibacterium is occasionally identified in infections involving prosthetic devices, especially shoulder devices [9,82,83]. (See "Overview of benign lesions of the skin", section on 'Epidermoid cyst' and "Pilonidal disease" and "Paronychia", section on 'Infectious agents' and "Overview of the evaluation and management of surgical site infection" and "Management of diabetic foot ulcers", section on 'Managing infection' and "Infectious complications of pressure-induced skin and soft tissue injury", section on 'Microbiology'.)

Deep soft tissue infections — Deep soft tissue infections likely to involve anaerobic bacteria include necrotizing fasciitis, synergistic necrotizing cellulitis, crepitant cellulitis, and gas gangrene. These infections involve the fascia, the muscle compartment formed by the enveloping fascia, or both. A relatively common feature of these infections is the low oxygen tension in tissues due to poor vascular supply; conditions in which this occurs include diabetes mellitus, injury (including surgery), and trauma. Major pathogens in these deep infections include group A beta-hemolytic streptococci, Clostridia, and combinations of aerobic and anaerobic bacteria [73,76,84,85] (table 2). (See "Necrotizing soft tissue infections" and "Clostridial myonecrosis".)

Human and animal bites — Human bites, and to a lesser extent animal bites, often involve anaerobic bacteria (table 2). These types of infections can arise from the oral flora of the biter or from the flora of the adjacent skin of the bite recipient. The clenched fist injury is the equivalent of a human bite in the sense that the infection involves the flora of the mouth of the person struck or the skin flora of the person delivering the punch [79]. (See "Human bites: Evaluation and management" and "Animal bites (dogs, cats, and other mammals): Evaluation and management".)

Central nervous system infections — Pyogenic intracranial infections that commonly involve anaerobic bacteria include cerebral, epidural, and subdural abscess [86,87]. The majority of brain abscesses are polymicrobial (eg, aerobic, anaerobic and/or facultative streptococci, staphylococcal species, other anaerobes, gram-negative rods) and the microbiology depends on the primary source for the infection (eg, complication of sinusitis, hematogenous, post-neurosurgery) (table 2). Isolation of Streptococcus anginosus (also known as Streptococcus milleri) suggests likely bacteremia with disseminated abscess formation [88].

Since meningitis rarely involves anaerobes, growth of these bacteria in cerebrospinal fluid (CSF) cultures suggests a parameningeal collection, shunt infection, or contamination. (See "Pathogenesis, clinical manifestations, and diagnosis of brain abscess" and "Infections of cerebrospinal fluid shunts" and "Intracranial epidural abscess".)

Bacteremia — Over time, anaerobes have accounted for about two to six percent of blood culture isolates from patients with clinically significant bacteremia. Actual frequency of detecting anaerobic bacteremia has differed geographically (eg, <0.5 to 10 percent), potentially related to the patient populations, underlying comorbidities, and local ability to isolate and identify anaerobes from blood cultures [6,89-91]. The value of anaerobic blood cultures has long been debated based on lower microbial yield and the idea that clinical suspicion for anaerobic infections was accurate. However, available data refute both these points and we continue to obtain anaerobic cultures as part of routine blood cultures [89,92,93]. The B. fragilis group, Clostridia, and anaerobic cocci are the most common isolates with typically B. fragilis sensu stricto considered to be the most common anaerobic bacterium isolated from blood culture specimens [5-7,94] (table 2). Anaerobic bacteremia crude mortality can be as high as approximately 25 percent, dictated by comorbidities, presence of sepsis, and/or adequacy of source control [89,94]. Isolated bacteremia is rare and should prompt further evaluation for the site of infection. Endocarditis caused by anaerobic bacteria is extremely rare and echocardiograms are generally not warranted in cases of isolated anaerobic bacteremia, except in cases of high suspicion (eg, persistent bacteremia despite appropriate therapy). Eggerthella lenta is a recognized cause of anaerobic bacteremia associated with intra-abdominal infections, resistance to piperacillin-tazobactam, and increased mortality [8].

Although anaerobic bacteremia accounts for a small percent of clinically significant bacteremia, B. fragilis sensu stricto and B. fragilis group bacteremia contributes significantly to morbidity and mortality [7,94,95]. As an example, a case-control study matched patients with B. fragilis group bacteremia to control patients without bacteremia but with the same principal diagnosis or the same major surgical procedure [94]. Patients with B. fragilis group bacteremia had a significantly higher mortality rate compared with non-bacteremic patients (28 versus 9 percent; attributable mortality rate of 19 percent). Further, a prospective multicenter, observational study confirmed high mortality associated with Bacteroides bacteremia especially if inactive therapy was administered (45 versus 16 percent, inactive versus active therapy, p = 0.04) [7]. In this report, antimicrobial susceptibility reliably predicted outcome of Bacteroides bacteremia.

DIAGNOSIS

When to suspect an anaerobic infection — Anaerobic infections should be suspected if any of the following clinical clues are present (table 3):

●Putrid drainage from wound – Putrid odor of wound discharge is considered diagnostic of anaerobic infection. However, it may be a relatively late feature and is only seen in approximately one-third to one half of patients [96,97]. The chemical basis for the odor is not well established, but it presumably reflects the metabolic products of anaerobic bacteria.

●Polymicrobial flora on Gram stain – Anaerobic infections often involve polymicrobial flora, so a Gram stain with multiple different morphotypes should raise suspicion for an anaerobic infection.

●Infection adjacent to mucosal surfaces that normally harbor anaerobic flora – Infections involving the upper respiratory tract, gastrointestinal tract, or the female genital tract often include anaerobes (table 2). (See 'Role in the human microbiome' above and 'Sites of infection' above.)

●Presence of abscess, gas, or tissue necrosis – Anaerobes are often associated with tissue necrosis and abscess formation. They are frequently isolated from abscesses at virtually all anatomic sites, including cerebral, dental, peritonsillar, lung, intra-abdominal, tubo-ovarian, prostatic, and some cutaneous abscesses. Gas in the tissue is another clue to the presence of anaerobic bacteria, but it is not considered diagnostic because some aerobic bacteria also produce gas (eg, Streptococcus pyogenes) [37]. In addition, gas detected by radiography or scanning techniques may also be due to air introduced during irrigations or other manipulations, such as the release of carbon dioxide or oxygen with hydrogen peroxide.

●Classic features of toxin-mediated and/or tissue destructive clostridial syndromes – Toxin-mediated and/or tissue destructive clostridial syndromes include tetanus, botulism, Clostridium perfringens food poisoning, gas gangrene, Clostridioides difficile-induced diarrhea or colitis, and enteritis necroticans. Clostridial species are also known to produce a "leukemoid reaction" causing a markedly elevated white blood cell (WBC) count. Clinical features consistent with any of these syndromes or a marked leukocytosis (WBC >30,000 cells/microL) should prompt suspicion for a Clostridial infection. (See 'Pathogenesis' above.)

If anaerobes are suspected, empiric therapy against anaerobes should be administered given the difficulty of identifying and isolating anaerobes on culture (table 4). (See 'Antimicrobial therapy' below.)

Microbiologic diagnosis — Microbiologic diagnosis of anaerobic bacteria is difficult due to a variety of reasons. Since anaerobic bacteria are part of normal human flora at virtually all anatomic sites, it is often problematic to determine when these bacteria represent true pathogens or merely commensals. Determining whether an anaerobe is a true pathogen is based on the recovered anaerobe, its known pathogenic potential, dominance of anaerobes in the sample, and the specimen source. For example, B. fragilis is known to cause life-threatening infections and should be considered a pathogen in the appropriate clinical setting (eg, abscess) but is also a frequent, low abundance member of the colon microbiota in healthy individuals.

Optimal clinical specimens for anaerobic culture are normally sterile body fluids or tissue specimens from sterile sites. Swab specimens should be avoided [1,81,98]. However, the presence of anaerobic normal flora on many mucosal sites makes obtaining a sterile specimen difficult. Using anaerobic transport media, maintaining specimens at room temperature (which helps to preserve anaerobe viability better than refrigeration), collecting samples prior to antibiotic administration, and rapidly transporting the samples to the laboratory all assist in anaerobe isolation. In certain circumstances, using techniques that allow acquisition of a more sterile specimen can be helpful to reduce the risk of contamination (eg, culdocentesis, laparoscopy).

Another obstacle to microbiologic diagnosis is the fastidious growth characteristics of anaerobes, making these organisms difficult to grow in clinical microbiology laboratories. Gram stain of exudate can provide important clues to the presence of anaerobic bacteria. Anaerobes vary in size and shape and are often pleomorphic. The Gram stain may show morphologic features suggestive of anaerobes, such as Bacteroides spp (small, delicate gram-negative bacilli), Fusobacterium nucleatum (pleomorphic fusiform bacteria with pointed ends), Fusobacterium necrophorum (long, "ropy" gram-negative bacilli), and Clostridia (large "box car"-like gram-positive bacilli). By contrast, Peptostreptococcus spp cannot be distinguished from aerobic gram-positive cocci by Gram stain appearance. (See 'Microbiology' above.)

TREATMENT

Antimicrobial therapy

Rationale for empiric treatment — Most anaerobic infections are treated empirically, since the process of recovery and in vitro testing is often delayed and limited in availability. Empiric therapy is generally effective when chosen based on knowledge of the bacteriology at the apparent source of the infection (eg, colonic perforation, ascending infection from the vagina such as PID, extension of infection from the sinus to the CNS) and the understanding of strain- and geography-dependent antimicrobial resistance patterns among anaerobic bacteria (table 4 and table 2) [1,99] (see 'Sites of infection' above).

Antibiotic selection — Effective antibiotics against anaerobic infections are important to optimize the patient's morbidity and mortality [7,100,101]. Management of mixed aerobe/anaerobic infections requires administration of antimicrobial agents active against each component of the infection.

●Preferred antibiotics – The following (classes of) antibiotics are the most active antibiotics against most anaerobic bacteria (including B. fragilis) and are preferred for the treatment of anaerobic infections (table 4) [10,11,99,102]:

•Metronidazole (very effective against gram-negative anaerobic bacteria and less reliable against gram-positive anaerobic bacteria)

•Beta-lactam-beta-lactamase inhibitor combinations (amoxicillin-clavulanic acid, ampicillin-sulbactam, piperacillin-tazobactam, meropenem-vaborbactam, imipenem-cilastatin-relebactam)

•Carbapenems

Both beta-lactam-beta-lactamase inhibitor combinations and carbapenems are very effective against both gram-positive and gram-negative anaerobic bacteria. However, ceftazidime-avibactam and ceftolozane-tazobactam do not have activity against many anaerobes, including B. fragilis and should not be used for the treatment of anaerobic infections.

The management of anaerobic infections for given infectious syndromes is discussed separately in the specific topic for each infectious syndrome.

●Antibiotics with limited activity – While clindamycin and moxifloxacin retain some activity against oral anaerobes, neither antibiotic is sufficiently effective in serious infections. When used, they should be used with caution given the varied susceptibility patterns.

Overall, penicillins, cephalosporins, and macrolides show limited activity against anaerobes due to widespread resistance and should not be used to empirically treat anaerobic infections. Up to 51 percent of all anaerobic bacteria isolated in bloodstream infections have been reported as resistant to penicillin including 20 percent of Clostridia species and 10 percent of gram-positive anaerobic cocci [103]. General in vitro susceptibility patterns of various antibiotics against anaerobes are summarized in the table (table 4). Multidrug resistance in anaerobes can occur [101,104-106].

The most important factor for the clinician to consider when initiating antibiotic therapy of anaerobes is the possibility of antibiotic resistance. (See 'Antimicrobial resistance' below.)

Many factors impact antibiotic resistance including local (hospital) and regional geography (country), antibiotic use patterns, site of infection and even the exact anaerobic species, although the latter will not be available at the time of empiric initiation of antibiotic therapy. Approaches to infection control and antimicrobial and diagnostic stewardship within hospitals can also impact the prevalence of anaerobic antimicrobial resistance. However, because local information may be limited, empiric antibiotic selection for treatment of anaerobic infections is often based on susceptibility test results from sentinel laboratories. If local rates of antimicrobial resistance are available, this information should guide clinicians in empiric antibiotic selection (see 'Antimicrobial resistance' below). Anticipated anaerobic pathogens for each site of infection are discussed elsewhere. (See 'Sites of infection' above.)

Other considerations for the clinician in selection of empiric antibiotic therapy for anaerobic infections include the preferred mode of administration (intravenous versus oral) for the clinical context and the location of the anaerobic infection (ie, specific site of infection). Although 'above versus below the diaphragm' is a classic construct in considering the selection of antimicrobial agents to treat anaerobic infections, the increasing resistance of oral anaerobes and B. fragilis group bacteria to antibiotics now limit its clinical usefulness.

Despite the difficulties of in vitro testing as described above, a clear correlation has been noted between antimicrobial in vitro activity and patient survival after treatment for monomicrobial and polymicrobial anaerobic bacteremia or tissue site infections [7,100,101]. This was illustrated, for example, in a prospective observational study of 128 patients with Bacteroides spp bacteremia [7]. The mortality rate, clinical failure, and microbiologic persistence by in vitro testing was higher among patients who received inactive therapy than in patients who received active therapy (mortality: 45 versus 16 percent; clinical failure: 82 versus 42 percent; microbiologic persistence: 22 versus 12 percent). (See 'Antimicrobial therapy' above.)

Antimicrobial resistance — Antibiotic resistance among anaerobic bacteria varies widely among different geographic regions and institutions, but is increasing worldwide (table 4) [10,107-113].

Multiple mechanisms of antimicrobial resistance exist in anaerobes including efflux pumps, porin loss, nim genes (especially important to metronidazole resistance in B. fragilis group organisms but also associated with resistance to other antibiotics) and other antimicrobial resistance genes (eg, cfiA [carbapenem], cepA/cfxA [beta-lactams], ermF [macrolide-lincosamide-streptogramin], tetQ [tetracycline], gyrA/gyrB/parC [fluoroquinolone]). These antimicrobial resistance genes mediate production of beta-lactamases, drug degradation, and 23S rRNA methylation as well as decrease affinity of antibiotics for bacterial target molecules. There are other mechanisms of antimicrobial resistance, but they still lack molecular definition. Both chromosomal and transferable plasmid-mediated resistance occur in anaerobes.

Bacteroides group members, and especially B. fragilis, are the most common anaerobes isolated from human infections and are more likely to be resistant to commonly used antimicrobial agents. Over time, the most significant changes have been with major reductions in the in vitro activity of beta-lactams (eg, penicillins, cefoxitin, cefotetan), clindamycin, and moxifloxacin against Bacteroides group members (especially B. fragilis). In contrast, resistance to metronidazole, beta-lactam-beta-lactamase inhibitor combinations, and carbapenems remains infrequent in most of the world.

●Clindamycin – Clindamycin was once a preferred antimicrobial agent for anaerobic infections but resistance can approach 60 percent (range 20 to 60 percent) across the globe inclusive of B. fragilis and B. fragilis group strains [10,108,109,111,112].

●Moxifloxacin – Resistance to moxifloxacin, the only fluoroquinolone with predicted anti-anaerobic activity, is now substantial (approximately 20 to 60 percent resistant). Thus, use of moxifloxacin alone for therapy of anaerobic infections is not recommended.

●Penicillins and cephalosporins – Only 40 to 50 percent of all anaerobes remain sensitive to penicillins [103,112] including <3 percent of B. fragilis or B. fragilis group strains [108,109,112]. In contrast, beta-lactam beta-lactamase inhibitors remain effective treatment options for anaerobic infections with >90 percent of all anaerobes tested susceptible [112]. However, susceptibility varies within the Bacteroides spp with B. fragilis being the most susceptible [10,108]. Parabacteroides (formerly considered part of the Bacteroides group) and other Bacteroides spp have been reported as only approximately 80 percent susceptible to beta-lactam beta-lactamase inhibitors [112].

●Carbapenems – Among beta-lactam agents, the carbapenems exhibit the broadest anaerobic organism coverage [111]. Ertapenem, meropenem, and imipenem are all generally equally active (>95 percent susceptible) against anaerobes including the majority of B. fragilis or B. fragilis group strains [10,111,112] although some studies have reported higher rates of resistance [111,112].

●Metronidazole – Metronidazole remains one of the most effective agents against anaerobic infections (>90 percent of all anaerobes susceptible) [112] and, in particular, is very effective treatment for gram-negative anaerobic bacteria. In the most recent United States-based multicenter surveillance of antimicrobial resistance among Bacteroides and Parabacteroides species, no resistance was detected in 779 isolates tested [10]. Surveys in other locales have yielded similar results (99 to 100 percent susceptible) [109,112]. Importantly, while metronidazole resistance is associated with nim genes, presence of a nim gene does not invariably lead to resistance [10,108]. Nonetheless, metronidazole resistance has been reported in B. fragilis and B. fragilis group strains [10,101,104,105,111].

●Multidrug resistance – Multidrug-resistant (MDR) B. fragilis have been reported in the United States; the isolates from these cases were resistant to metronidazole, carbapenems, piperacillin-tazobactam, and clindamycin, among other agents [104,105].

Adjunctive source control — Anaerobic infections are frequently associated with abscess formation. This feature along with the often polymicrobial makeup (ie, mixed infections with aerobes) of anaerobic infections makes source control (eg, draining pus, debriding necrotic tissue, improving tissue oxygenation) a critical consideration in the clinical approach to anaerobic infections. Antibiotic therapy is often adjunctive to source control in the cure of anaerobic infections [107,114].

Duration of therapy — The duration of therapy is based on a multitude of clinical factors, including severity and site of infection, presence of abscesses, and whether the source of infection has been controlled. Details on duration of therapy for given infectious syndromes are discussed in the specific topic reviews.

SUMMARY AND RECOMMENDATIONS

●Microbiology – Anaerobic bacteria are fastidious, slow-growing organisms that require low-oxygen conditions for growth and can be difficult to grow in clinical microbiology laboratories. The major clinically significant anaerobes are summarized in the table (table 1). (See 'Microbiology' above.)

●Sites of infection – Anaerobes are part of the normal skin, genital, oral, and gastrointestinal tract (especially the colon) flora. Infections often originate due to a break in the mucosal barrier at these sites and spread to other sites contiguously. Infections range from relatively minor skin and oral infections (eg, pharyngitis and gingivitis) to life-threatening invasive infections (eg, bacteremia, distant focal suppurative metastatic infections involving the brain, deep neck space, or muscle and fascia) and are often polymicrobial (table 2). (See 'Sites of infection' above.)

●Diagnosis – Clinical features that should prompt suspicion for an anaerobic infection are presented in the table (table 3). Most clinical laboratories are able to identify the most clinically important anaerobes (table 1). (See 'When to suspect an anaerobic infection' above.)

●Management

•Empiric therapy – If anaerobes are suspected, empiric therapy against anaerobes known to cause infection at the apparent source of infection should be administered given the difficulty of identifying and isolating anaerobes on culture (table 2). (See 'Rationale for empiric treatment' above.)

•Antimicrobial therapy – General in vitro susceptibility patterns of various antibiotics against anaerobes are summarized in the table (table 4).

-Effective – Metronidazole, beta-lactam-beta-lactamase inhibitor combinations, and carbapenems are the most active antibiotics against B. fragilis as well as most other anaerobic bacteria, based on susceptibility test results from sentinel laboratories (table 4). Ceftazidime-avibactam and ceftolozane-tazobactam do not have activity against many anaerobes, including B. fragilis, and should not be used for the treatment of anaerobic infections. (See 'Antibiotic selection' above.)

-Insufficiently effective – While clindamycin and moxifloxacin retain some activity against oral anaerobes, both antibiotics should be used with caution given the varied susceptibility patterns and should be avoided in serious infections (table 4). Penicillins, cephalosporins, and macrolides show limited activity against anaerobes due to widespread resistance. (See 'Antibiotic selection' above and 'Antimicrobial resistance' above.)

•Adjunctive source control – Anaerobic infections are frequently associated with abscess formation. This feature along with the often polymicrobial makeup (ie, mixed infections with aerobes) of anaerobic infections makes source control (eg, draining pus, debriding necrotic tissue, improving tissue oxygenation) a critical consideration in the clinical approach to anaerobic infections. (See 'Adjunctive source control' above.)

•Duration of antimicrobial therapy – The duration of therapy is based on a multitude of clinical factors, including severity and site of infection, presence of abscesses, and whether the source of infection has been controlled. (See 'Duration of therapy' above.)

ACKNOWLEDGMENT — The editorial staff at UpToDate acknowledge John G Bartlett, MD, who contributed to an earlier version of this topic review.