Controversies in anaerobic infections in childhood

Ztzhak Brook, M.D., M.Sc., is Professor of Pediatrics and Surgery at the Uniformed Services University of the Health Sciences and Consultant in Infectious Diseases at the Naval Medical Center, National Capital Region, Bethesda, Maryland, and Walter Reed Army Medical Center in Washington, D.C. He is also a Senior Investigator at the Armed Forces Radiobiology Research Znstitute in Bethesda and the Chairman of the Anti-infective Drug Advisory Committee of the Food and Drug Administration. Dr. Brook is a graduate of the Hebrew University School of Medicine, Jerusalem, Israel, and received his training in pediatrics at Kaplan Hospital, Rehovot, Israel. He completed. an M.S. program in the Weitzman Institute, Rehovot, Israel, and the Tel-Aviv University. He received his training in infectious diseases at UCLA, and served on the faculty of Pediatrics and Infectious Diseases at the University of California, Irvine, Medical Center and Children’s Hospital National Medical Center, Washington, D.C. His research interests relate to the role of anaerobic bacteria in pediatric infections. Curr

Probl

Pediatr,

October

1987

CONTROVERSIES INFECT’IONS

IN ANAEROBIC IN CHILDHOOD

INTRODUCTION

Anaerobes have been isolated with increased frequency from many infections in children.l Their increased rate of isolation has alerted many pediatricians to their importance in various infectious processes, where they have thus far not been recognized. Several types of infections frequently involve anaerobic bacteria.” 3 These include infections of the central nervous system, especially brain abscess and subdural empyema; a variety of orofacial infections and pleuropulmonary infections; especially lung abscess, aspiration pneumonia, necrotizing pneumonia, and empyema. Intra-abdominal infections, for example, secondary peritonitis, intra-abdominal abscess, appendicitis, and diverticulitis, commonly involve anaerobic bacteria. Infections of the female genital tract, such as salpingitis, pelvic abscess, and endometritis, also involve anaerobic bacteria. A variety of soft tissue infections, such as clostridial myonecmsis, synergistic necrotizing cellulitis, and necrotizing fasciitis, are caused by anaerobic bacteria. Anaerobic bacteremia is a frequent complication of these infections. However, in contrast to adults, where these organisms are often isolated in patients with chronic debilitating diseases, in children these organisms are more often isolated in normal individuals. Anaerobes predominate more often in children with infections of the upper respiratory tract, as these are more commonly seen in the young. Neonatal anaerobic infections are also unique to the pediatric age group. Recent data have also implicated some anaerobic bacteria in the increased failure rate of penicillins in the therapy of many infections through the production of the enzyme beta-lactamase.4 This indirect pathogenic role of anaerobic bacteria may have direct implication on the management of different pediatric infections such as recurrent tonsillitis, chronic otitis media, skin and soft tissue abscess, pulmonary abscess, and other types of polymicmbial infections. Despite increased awareness of the importance of anaerobic bacteria in many infections in children, controversies exist about their Curr

Probl

Pediatr,

October

1987

563

role. Doubt is often raised whether these organisms represent and whether therapy should be also directed at true pathogens, their elimination. Their importance in “shielding” penicillin-susceptible pathogens by virtue of production of beta-lactamase, and the implication of that phenomenon on management of polymicrobial infections, is still under evaluation5 Clinicians often wonder when they should obtain culture for the recovery of anaerobic bacteria and when and if susceptibility testing should be done on these isolates.6 The following review will present these controversial issues in the perspective of current information and will discuss the evidence in support of the role of anaerobes in pediatric infections. It will focus on their direct and indirect role in respiratory, abdominal, pelvic, and skin and soft tissue infections, and implications for the management of these infections.

exact

PROPER VS. TRANSPORT

1MPROPER TECHNIQUES OF SPECIMENS

OF

COLLECTION

AND

The perception that anaerobes have little or no role in many infections originates from the fact that many past studies did not attempt to identify them, or utilized improper methods for collection of specimens for anaerobes. It is therefore essential to carefully assess any such study for its methodological properties prior to judging its ability to determine the role of anaerobes in an infectious process. Multiple examples of differences in the rate of recovery of anaerobic bacteria, between studies that used proper and those that utilized improper techniques, can be found. Earlier studies of chronic otitis media7 and human and animal bites8 that did not employ a method for anaerobes found these organisms in a small number of cases. However, when better techniques were utilized, anaerobes were recovered in the majority of the casesgJ1’ Since anaerobes may invade any body site, and they have been recovered in a variety of infections in children, their potential role in an infectious site should be assessed individually. The prevalence of anaerobic bacteria in an infection is a major factor in deciding which clinical specimens should be processed for anaerobes. The proper management of anaerobic infection depends on appropriate documentation of the bacteria causing the infection. Without such an approach the patient may be exposed to inappropriate, costly, and undesired antimicrobial agents with adverse side effects. Anaerobic infections present special bacteriologic problems not encountered in other types of infections, and such problems may make the therapeutic approach even more difficult. Generally, bacteriologic results will not be available as quickly as in aerobic infections, particularly if the infection is mixed (as are more than one half of the cases). Some laboratories may fail to recover certain or all of 564

Curr

Probl

Pediatr,

October

1987

the anaerobes present in a specimen. This situation can occur particularly when the specimen is not promptly put under anaerobic conditions for transport to the laboratory. If care is not taken to avoid contamination of the specimen with normal flora, anaerobes may be recovered that have little to do with the patient’s illness. As all laboratories are not equipped to accurately identify anaerobes, presumptive results may be very misleading. Appropriate cultures for anaerobic bacteria are especially important in mixed aerobic and anaerobic infections. Techniques or media that are inadequate for isolation of anaerobic bacteria, either because of a lack of an anaerobic environment or because of an overgrowth of aerobic organisms, can mislead the clinician to assume that the aerobic organisms recovered are the only pathogens present in an infected site, therefore causing the clinician to direct therapy toward those aerobic organisms only. The nature of the various organisms in a mixed infection will also influence the choice of drugs. Drugs active against anaerobic bacteria may be quite inactive against the accompanying aerobic or facultative organisms. When mixed infections involve several organisms, two, three, or more drugs may be required to provide effective coverage for each of the organisms in the mixture. Since anaerobic bacteria frequently can be involved in various infections, ideally, all properly collected specimens should be cultured for these organisms. Special efforts should be made by the physician to isolate anaerobic organisms in infections in which these organisms are frequently recovered, such as abscesses, wounds in and around the oral and anal cavities, chronic otitis media and sinusitis, and aspiration pneumonia, among others. The most acceptable documentation of an anaerobic infection is thorough culture of anaerobic microorganisms from the infected site. Three elements requiring the cooperation of the physician and the microbiology laboratory are essential for appropriate documentation ot anaerobic infections: collection of appropriate specimens, expeditious transportation of the specimen, and careful laboratory processing. Collection of Specimens.-Specimens must be obtained free of contamination so that saprophytic organisms or normal flora are excluded and culture results can be interpreted correctly. Because indigenous anaerobes are often present in large numbers on the surface of skin and mucus membranes, even minimal contamination of a specimen with the normal flora can give misleading results. On this basis, specimens can be designated according to their acceptability for anaerobic culture. Materials appropriate for anaerobic cultures should be obtained using a technique that bypasses the normal flora. Unacceptable or inappropriate specimens can be expected to yield normal flora also, and therefore have no diagnostic value. Curr

Probl

Pediatr,

October

1987

565

Examples of these specimens include coughed sputum, bronchoscopy aspirates, gingival and throat swabs, feces, gastric aspirates, voided urine, and vaginal swabs (Table 1). Exceptions to these guidelines can be made if in certain instances the clinical condition warrants such a culture. An example is the use of selective media to detect a possible pathogen only, such as Clostridium d@ciZe in stool obtained from a patient with colitis. Acceptable specimens include blood, aspirates of body fluids (pleural, pericardial, cerebrospinal, peritoneal, and joint fluids), urine collected by percutaneous suprapubic bladder aspiration, abscess contents, deep aspirates of wounds, and specimens collected by special techniques such as transtracheal aspirates or direct lung puncture (Table 2). Direct needle aspiration is probably the best method of obtaining a culture, whereas use of swabs is much less desirable. Specimens obtained from normally sterile sites may be collected after thorough skin decontamination (i.e., site of blood collection, spinal joint, or peritoneal fluids). Cultures of coughed sputum and specimens obtained from bronchial brushing or bronchoscopy are generally contaminated with normal oral and nasal aerobic and anaerobic flora and are therefore unsuitable for culture. Because the trachea below the thyroglossal membrane is sterile in the absence of pulmonary infection, transtracheal aspiration (‘ITA), done below this site, is a reliable procedure for obtaining suitable culture material for the diagnosis of pulmonary infection.l’, I2 An alternative procedure is direct lung puncture. These procedures, when performed by experienced operators, yield important data, and the complication rates are very low. However, TTA is not usually recommended in patients with severe hypoxia, hemorrhagic diathesis, or severe cough.13 Rare complications, such as hypoxia, bleeding, subsequent emphysema, or arrhythmia, have rarely been reported in adult patients.14 Transtracheal aspiration has also been successfully utilized in the diagnosis of aspiration pneumonia and lung abscess in childrenl’ Cultures obtained through TTA contain fewer pathogens than cultures of expectorated sputum. TABLE Specimens

1. That

Should

Not

be Cultured

for /maerobes

Feces or rectal swabs Throat or nasopharyngeal swabs Sputum or bmnchoscopic specimens Routine or catheterized urine Vaginal or cervical swabs Material from superficial wound or abscesses not collected to exclude surface contaminations Material fmm abdominal wounds obviously contaminated (e.g., open fistula)

566

Curr

Probl

properly with

Pediatr,

feces

October

1987

TABLE Specimens

2. Appropriate

for

Anaerobic

Culture

All normally sterile body fluids other than urine (e.g., blood, pleural fluid, joint fluid1 Urine obtained by suprapubic bladder aspiration Percutaneous transtracheal lung puncture Culdocentesis fluid obtained of the vagina Material obtained Material obtained wounds

fmm from

closed sinus

aspiration after

or direct decontamination

abscesses tracts or draining

Side effects of this procedure in children include and in rare instances, subcutaneous emphysema.

mild

hemoptysis,

Transport of Specimens.-The ability to recover anaerobes is influenced by the care applied to transport and laboratory processing of specimens. Unless proper precautionary measures are taken during collection, transport, and laboratory processing, pronounced changes can occur in the aerobic and anaerobic microbial population of a clinical specimen.15 Sensitivity to oxygen causes some obligate anaerobes to die rapidly upon exposure to air. In clinical samples, obligate anaerobes can also be overgrown by facultative anaerobes unless processed rapidly after collection. The organisms, therefore, have to be protected from the deleterious effects of oxygen during the time between the collection of the specimen and their inoculation into the proper anaerobic medium in the microbiology laboratory. Failure to take proper precautions can result in misleading data, which indirectly may be detrimental to the patient.“15-17 Anaerobes vary in the conditions they require for survival. Some organisms are classified as moderate and others as fastidious in accordance with their oxygen sensitivity.18 Among the moderate group (those capable of growing in an oxygen concentration of 2% to 8%) are Bacteroides fragilis, Bacteroides oralis, Bacteroides melaninogenicus, Fusobacterium nucleatum, and Clostridium perfi-ingens. Some fastidious anaerobes will grow in 0.5% oxygen, and some are extremely oxygen-sensitive, such as some strains of B.fragiZis and peptostreptococci.’ Low oxidation-reduction potential is another basic requirement for growth of certain anaerobic bacteria such as Bacteroides vulgatus and Clostridium sporogenes.lg Such conditions usually exist in areas where anaerobes are present as part of the normal flora and at infected sites. The implication of these observations is that specimens must be carefully and rapidly handled in both transport and processing to ensure good recovery of anaerobes. The specimens should be placed as soon as possible after their Cm-r

Probl

Pediatr,

October

1987

567

collection into an anaerobic transporter. Aspirates of liquid specimen or tissue are always preferred to swabs, although systems for the collection of all three culture forms are commercially available. Several versions of anaerobic transporter are commercially available. These transport media are very helpful in preserving the anaerobes until the time of inoculation. Liquid specimens may be inoculated into a commercially available anaerobic transport vial that is devoid of oxygen and sometimes contains an indicator. A plastic or glass syringe and needle may also be used for transport. After the specimen is collected and all air bubbles expelled from the syringe and needle, the needle tip should be inserted into a sterile rubber stopper. Because air gradually diffuses through the wall of a plastic syringe, no more than 30 minutes should elapse before the specimen is processed. This inexpensive transport device for liquid specimens is especially useful in the hospital, for instance, where the specimen can be rapidly transported to the microbiologv laboratory? Swabs may be placed into sterilized tubes containing carbon dioxide or anaerobically prereduced, sterile Cary-Blair semisolid media. A preferred method utilizes a swab that has been prepared in a prereduced anaerobic tube. Tissue specimens or swabs can be transported anaerobically in an anaerobic jar or in a Petri dish placed in a sealed plastic bag made anaerobic by using a catalyzer. Most of the common and clinically important anaerobic bacteria are moderate anaerobes, as shown by the examination of various types of clinical specimens for anaerobes .‘O Because numbers and kinds of microorganisms in clinical materials vary widely, no transport device should be expected to give optimal protection for all anaerobes that may be encountered in specimens. Syed and Loesche’l reached this conclusion after studying the survival of human dental plaque flora in various transport media. Although some of the transport systems can support the viability of anaerobic organisms for up to 24 hours,22,23 all specimens should be transported and processed as rapidly as possible after collection to avoid loss of fastidious oxygen-sensitive anaerobes and the overgrowth of facultative bacteria. When delay in transportation is expected, refrigeration of the sample may prevent overgrowth of some organisms and preserve their distribution. Several investigators, however, have found refrigeration to be of little benefit.” Conclusions.-It is imperative that a physician treating a patient with a suspected anaerobic infection utilize appropriate methods of obtaining cultures of the infected site. This requires that the normal flora are bypassed in obtaining the culture and that an appropriate 56s

Curr

Probl

Pediatr,

October

1987

and rapid transport system ical data can be obtained formed. SHOULD SPECIES

ANAEROBIC LEVEL?

is used thereafter. only when these

BACTERIA

BE

Reliable microbiologprocedures are per-

IDENTIFIED

TO

A

’

Identification of an anaerobe to a species level is often cumbersome and expensive, and takes up to 72 hours. The decision of what level of speciation is adequate for identification of an anaerobic organism is often a controversial issue. The clinician often has to make such a decision. Occasionally, species identification of an organism will provide the diagnosis. Such is the case with C. diflcile in a patient with colitis, or Clostridium botulinurn in infants with botulism.” 3 However, since most anaerobes are endogenous, these are rarely epidemiologic reasons to do their complete identification. Identification of the B. fragilis group, that frequently causes bacteremia and septic complications, has significant prognostic value. Identification of an anaerobe is most helpful in determining what antibiotic to use against these species whose antibiotic susceptibility is predictable. Until the late seventies, virtually all clinical significant anaerobes, except B. fragilis group, were susceptible to penicillin.2’3 It was therefore unnecessary to extensively speciate the antibiotic susceptibility. However, in the last decade there have been significant changes, and now there is more variability in antimicrobial susceptibility patterns. These changes have necessitated more extensive speciation as well as antimicrobial susceptibility testing for some anaerobic bacteria. Organisms that should be identified are (1) isolates from sterile body sites lie., blood, cerebrospinal fluid, joint fluid; (2) an organism with particular epidemiologic or prognostic significance (i.e., C. d@ciZe); and (3) an organism with variable or unique susceptibility. SHOULD ANAEROBIC

THE

ANTIMICROBIAL BACTERIA BE

SUSCEPTIBILITY DETERMINED?

OF

The susceptibility of anaerobic bacteria to antimicrobial agents has become less predictable. Resistance to several antimicrobial agents by B. fi-agiZis group has increased over the past decade.24 A decrease in susceptibility to penicillin of C. per-j?ingens has also been noted.25 It is therefore impohant to perform susceptibility testing to isolates recovered from sterile body sites or those that are clinically significant and have variable or unique susceptibility. In addition to susceptibility testing per se, screening of anaerobic isolates (particularly Bacteroides sp.) for beta-lactamase activity may Curr

Probl

Pediatr,

October

1987

669

bacilli for be helpful. We routinely screen anaerobic gram-negative beta-lactarnase production using the nitrocefin disc.26 Such beta-lactamase screening of these isolates rapidly provides information regarding the penicillin susceptibility of these isolates. However, occasional bacterial strains may resist beta-lactam antibiotics through mechanisms other than the production of beta-lactamase. It must be recognized that routine susceptibility testing of all anaerobic isolates is extremely time-consuming and in many cases testing should be limited to those unncessary. Therefore, “routine” anaerobes isolated from blood cultures, bone and joint infections, serious otolaryngological infections, and central nervous system infections, as well as to anaerobic bacteria isolated in pure culture.z7 Antibiotics tested should include penicillin, a broad-spectrum penicillin, clindamycin, chloramphenicol, metronidazole, cefoxitin, and imipenem. WHAT ARE THE PREDOMINANT ANAEROBIC BACTERIA IN PEDIATRIC INFECTIONS, AND WHICH OF THEM SHOULD BE IDENTIFIED? The important anaerobes isolated from pediatric Peptostreptococcus sp., Clostridium sp., Bacteroides bacterium sp. (Table 3).

infections are sp., and Fuso-

Anaerobic Cocci.-The gram-positive anaerobic cocci most commonly isolated from clinical specimens are Peptostreptococcus magnus, Peptostreptococcus anaerobius, Peptostreptococcus intermedius, Peptostreptococcus asaccharolyticus and Peptostreptococcus prevotii. Veillonella parvula is the most common gram-negative coccus encountered. These organisms are part of the normal flora of the mouth, upper respiratory tract, intestinal tract, vagina, and skin. Anaerobic cocci have been recovered from children with subcutaneous abscesses” and burnszg around the oral and anal area, intraabdominal infections,30 decubitus u1cers,31 bacteremia,32 and brain abscesses.33’ 34 These organisms are predominant isolates in all types of respiratory infections, including chronic sinusitis,35 mastoiditis,36 acute and chronic otitis media,37’38 aspiration pneumonia,3g and lung abscesses.40 They generally are recovered mixed with other anaerobic or aerobic organisms, but in many cases they are the only pathogens recovered. Microaerophilic streptococci are of particular importance in chronic sinusitis35 and brain abscesses.33,34 There are essentially no unique anaerobic cocci that should be identified. These organisms are rather uniformly susceptible to penicillins and cephalosporins as well as to other antibiotics. Thus, these organisms may be identified solely as anaerobic gram-positive

570

Curr

Probl

Pediatr,

October

1987

of Anaerobic

Recoverv

'Frequency

of recovery

Bacteremia Central nervous system Head and neck Thoracic Abdominal Obstetric-gynecologic Skin and soft tissue

INFECTION

3.

TABLE

in anaerobic

infections:

1 1 3 2

3 2 3 3 2

2 = common

(34%-66Dhl;

3 = very common

1

2 1

1

1 2

3 1 2

0 0

3 3

0 0

DEW&V

ORALIS 1 2

EACTEROIDES BNWS EACTEROIDES

BACTEROIDES MEI.ANlNOGENICUS/

1 1

1

2

@JUP

RACTEROIDES FRAGKIS

0 = none; 1 = rare Cl%-33%);

1

1 1

SP.

1 2

CL.OSTR,DI”M

Patients*

SP.

in Pediatric

PEPTOSTKE~OCOCCU.5

Bacteria

l67%-100%)

1 1

1

3 3

1 1

SP.

FUSOBACTERIUM

cocci or anaerobic gram-negative they are slow-growing organisms ident@.

cocci, a fortunate situation since that may be somewhat difficult to

Clostridizzm Species.-The three Clostridium species found most frequently in clinical infections are C. peeingens, Clostridium ramosum and Clostridium innocuum. These three species, as well as others, have been recovered from the blood of children with gastrointestinal disease3” 32 and sickle cell disease.41 Clostridial strains have been recovered from specimens obtained from children with acute and chronic otitis media,38,42 chronic sinusitis and mastoiditis, 35236peritonsillar abscesses,43 peritonitis,30 neonatal conjunctivitis,44 and omphalitis .45 Although C. botulinurn usually is associated with food poisoning, wound infections caused by this organism are being recognized with increasing frequency. In newborns, C. botulinurn has produced a clinical picture of hypotonia, respiratory arrest, areflexia, ptosis, and poorly responding pupils. Botulism in infants is caused by a toxin produced by ingested spores of C. botulinurn that germinate in the bowel lumen. Toxigenic Clostridium butyricum has been recovered from patients with necrotizing enterocolitis,46P 47 but the role of this organism and other Clostridium sp. in necrotizing enterocolitis in the newborn is unclear. C. diflcile has been incriminated as the causative agent in antibiotic-associated diarrhea and colitis, as well as in spontaneous diarrhea in children .4884g However, their exact role in the latter is uncertain. C. per@ingens, C. butyricum and C. dz@YZe have been recovered from blood and/or peritoneal fluid of patients with, necrotizing enterocolitis46”0 and of infants with sudden infant death syndrome .47 Infections caused by Clostridium tetani are usually a result of contamination of wounds with soil containing C. tetani spores. The spores will germinate in devitalized tissue and produce the neurotoxin responsible for the clinical findings. C. tetani has been recovered from patients who present with otogenous tetanus.51 Identification of most of the clostridial species that are recovered from normally sterile sites may therefore provide assistance in clinical diagnosis. The possible use of antitoxins against several of these bacteria may also justify such effort. In addition to identifying C. perfringens and C. botulinurn as well as other toxin-producing species, it is now important to identify C. ramosum, since this organism is becoming increasingly resistant to clindamycin.52 The gram-positive non-spore-forming anaerobic rods are common isolates, but since they remain uniformly susceptible to penicillin, it is not necessary differentiate them beyond the genus level (i.e., Eubacterium sp. and Propionibacterium sp.). 572

Curr

Probl

Pediatr,

October

1987

Actinomyces Species.-Actinomyces israelii and Actinomyces naesZundii are normal inhabitants of the human mouth and throat (particularly gingival crypts, dental calculus, and tonsillar crypts), and are the most frequently isolated Actinomyces. These organisms have been recovered from intracranial abscesses,33 chronic mastoiditis,36 and aspiration pneumonia.39 Although Actinomyces are often present in mixed culture, they are clearly pathogenic in their own right and may produce widespread and devastating disease anywhere in the body.53 The lesions of actinomycosis occur most frequently in the tissues of the face and neck, lungs, pleura, and ileocecal regions. Bone, pericardial, and anorectal lesions are less common, but virtually any tissue may be invaded. The differentiation of actinomycetes from other anaerobic gram-positive rods has significant therapeutic implications. However, further speciation of actinomycetes is generally not necessary. Bacteroides Species.-Of all the organisms in the Bacteroidaceae family, B. fragilis group is isolated most frequently. These organisms are resistant to penicillins, mainly because of the production of betalactamase. Because of their presence in normal flora of the gastrointestinal tract, these organisms predominate in bacteremia associated with intra-abdominal infections,30’ 32 peritonitis, and abscesses following rupture of a viscus, and subcutaneous abscesses or burns near the anus .28J29 Although B. fragilis is not usually part of the normal oral flora, it can colonize the oral cavity of patients with poor oral hygiene or those who previously received antimicrobial therapy (especially penicillin). Following colonization of the oropharyngeal cavity, this organism can also be recovered from pediatric infections that originate in this area, such as aspiration pneumonia,3g lung abscesses,4o chronic sinusitis,35 chronic otitis media,37 brain abscesses,33 and subcutaneous abscesses or burns near the oral cavity.z8,2g B. fragilis can be recovered from infectious processes in the newborn. The newborn infant is at risk of developing these infections when born to a mother with amnionitis or premature rupture of membranes during the newborn’s passage through the birth canal, where B. fiagilis is part of the normal flora.54 B. fragilis has been recovered from newborns with aspiration pneumonia,55 bacteremia,32 omphabtis,45 and subcutaneous abscesses and occipital osteomyelitis following fetal monitoring.5” B. fragilis group organisms have become increasingly resistant to penicillin. However, in some areas B. fragiZis strains have become increasingly resistant to clindamycin as well as to other antibiotics. Differences in susceptibility to cephalosporins among members of the B. fragilis group has also bee noted.57 While most B. fragilis are susceptible to several of the newer cephalosporins, other members of the group are relatively resistant to these drugs, but susceptible Curr

Probl

Pediatr,

October

1987

573

to identify B. fiagilis as well to cefoxitin.57 It is therefore important as other members of the B. fragilis group. The Bacteroides melaninogenicus group, Bacteroides oralis and other Bacteroides sp. are part of the normal oral and vaginal flora, and they have emerged as the predominant Bacteroides species isolated from respiratory infections. These include aspiration pneumonia,3g lung abscess,4o chronic otitis media,37’ 38 and chronic sinusitis.34 These organisms have been recovered also from abscesses and burns around the oral cavity,2822g human and animal bites and other wounds,10 paronychia;’ urinary tract infections,5g and brain abscesses.33’ 34 They have also been isolated from children with bacteremia associated with infections of the upper respiratory tract.3z Organisms in the B. melaninogenicus group are important pathogens in periodontal disease and periodontal abscesses in children6’ Many Bacteroides sp., including B. fragilis, produce beta-lactamase, which enables them to resist penicillin and many cephalospotins. Previously, most B. melaninogenicus and B. oralis strains were considered to be susceptible to penicillin; within the last decade, however, penicillin-resistant strains have been reported with increasing frequency in children.61 The appearance of penicillin resistance among Bacteroides sp. has important therapeutic implications. The organisms, when present in a localized soft tissue infection, release beta-lactamase into the environment, thereby protecting not only themselves but also other pathogens that are penicillin-sensitive.5 Thus, penicillin therapy directed against a susceptible pathogen might be rendered ineffective by the presence of penicillinase-producing organisms. The phenomenon has been shown to be clinically significant in several mixed aerobic-anaerobic infections, such as recurrent tonsillitis,“’ chronic otitis media,63 and subcutaneous abscesses.4 Bacteroides disiens and Bacteroides bivius are the predominant Bacteroides isolated from female genital infections.64 An increase in resistance to penicillin has also been noticed by these organisms.27 Because of the increased resistance to penicillins of Bacteroides sp. other than B.fragiZis, it is important to identify these organisms. Whenever the microbiological laboratory is not able to identify the Bacteroides to their species level, a rapid test for detection of the production of beta-lactamase can provide the clinician with important useful information. The production of the enzyme by an isolate will mandate the use of an antimicrobial resistant to it. Fusobacterium Species.-The Fusobacterium sp. seen most often in clinical infections are Fzzsobacterium nucleatum, Fusobacterium necrophorum and Fzzsobacterium mortijkum. F. nucleatum is the predominant Fusobacterium species recovered from clinical specimens. It is often associated with infections of the mouth,43,62 lung,3g and brain.33’ 34 Because these organisms are part of the normal oral 574

Curr

Probl

Pediatr,

October

1987

and gastrointestinal flora, they are found in almost all types of infections in children, including bacteremia,3z meningitis associated with otologic disease,34 peritonitis following rupture of viscu~,~’ and subcutaneous abscesses and burns near the oral or anal orifices.28~2g Fusobacterium varium is generally resistant to penicillin and should be identified, but other fusobacteria need only be identified to the genus level. The recent description in Sweden of beta-lactamase production by other Fusobacterum sp. highlights the need to monitor the possible appearance of such strains in other countries.65 It is therefore advisable to perform a rapid beta-lactamase test in all Fusobacterium isolates. CONTROVERSIES

IN

SPECIFIC

INFECTIONS

The role of anaerobes in many infections in children is well established. These include infections such as intracranial abscesses, intraabdominal infections, bacteremias, osteomyelitis, and arthritis3 However, the exact significance of isolation of anaerobes in several of these infections is still controversial. The uncertainty regarding their role stems from either the paucity of studies documenting their presence (i.e., acute otitis media, pneumonitis, or cystic fibrosis), lack of studies utilizing animal models that demonstrate virulence, and difficulty in differentiating the anaerobes suspected of local virulence from those belonging to the normal flora (i.e., tonsillitis). A brief discussion of the major disputed clinical infections is presented. Necrotizing EnterocoZitis.-The role of bacteria in necrotizing enterocolitis (NEC) has been widely studied, and organisms such as Escherichia coli and KZebsieZZa pneumoniae have been isolated in various local outbreaks. A search of anaerobes as a cause of NEC started only after the demonstration of the role of C. d@ciZe in antibiotic-associated colitis.66 In contrast to the unique role of C. difltile in antibiotic-associated colitis in adults, several Clostridial sp., including C. d@ciZe, were associated with NEC. CZostridium sp., such as C. perfSngens,67 C. butyricum,46847 and C. d@ciZe6’ have been isolated from several nursery outbreaks. Evaluation of C. diflcile was particularly difficult since this organism is a frequent colonizer of the gastrointestinal tract of newborns6’ and often does not elaborate toxins in newborns.70 The hypoxia and circulatory disturbances in small premature infanv at risk of NEC may lead to ischemia that may promote the growth of Clostrium sp. However, the exact role of CZostridium sp. in NEC awaits further studies. OmphaZitis.-Anaerobic colonize the umbilical CWT

Probl

Pediatr,

October

bacteria are among the organisms that stump after delivery. These bacteria may in1987

575

vade the wound and cause local or systemic infection. The polymicrobid aerobic-anaerobic etiology of neonatal omphalitis has been Anaerobes (predominantly Bacteroides) recently demonstrated.45 were isolated in 39% of the infected umbilical cods. Although the fom-smelling secretions of an infected umbilical cord are a typical manifestation of an anaerobic iufection, it is possible that the anaerobic bacteria are only colonizers. Culturing of an infected cord should therefore be done in methods that avoid contamination by skin flora. Skin and Sofl Tissue Infections.-The most common etiologic agents in skin and soft tissue infections are Staphylococcus aureus and Group A beta-hemolytic streptococci (GABHS). However, anaerobes have been isolated from such infections that are adjacent to mucus surfaces where they form part of the normal flora,” and S. azzreus has been isolated from all body sites. Anaerobes were recovered more often from abscesses in and around the rectal or vulvovaginal area (B. $+agiZis and Clostridium sp,), around the oropharynx (head and neck) and fingers (paronychiaIs8 and from human bites (B. melaninogenicus group, B. oralis)” (Table 4). Similar distribution of organisms were recovered from decubitus ulcers3’ and burn wounds.” Decubitus ulcers and burns in and around the oropharynx harbored oral flora anaerobes, and those close to the rectal or vulvovaginal area contained gastrointestinal or vaginal flora. Controversies regarding the exact role of these organisms may exist, especially when the cultures that are obtained are from the surface of the wound. However, biopsies (in burns) or properly collected pus specimens (wounds and abscesses) are reliable for providing accurate diagnosis. Conjunctivitis.-Although many organisms can cause conjunctivitis, only a few studies have investigated the role of anaerobes in this TABLE Predominant Children*

4. Organisms

Isolated

From

Staphylococcus aureus Streptococcus sp. Enteric bacteria Peptostreptococcus sp. Clostridium sp. Bacteroides fragilis group Bacteroides melaninogenicus

570

Abscesses

NO. ORGANISMS (‘?‘o OF PATIENTS)

BACTERIA

‘Data from Brook 1981; 67:891.

Subcutaneous

I, Finegold

PREDOMINANT

76 (43%) 38 (22%)

group SM: Aerobic

27 62 6 29 24 and anaerobic

Sites, in 176

SITE OF ISOLATION

Extremities, trunk Extremities, trunk Rectal, vulvovaginal Rectal, vulvovaginal, finger Rectal Rectal, vulvovaginal Head, fingers, vulvovaginal

(15%) (35%1 (3%) (16%) (14%) bacterial

at DilTerent

flora of burns

Curr

Probl

in children.

Pediatr,

October

Pediatrics

1987

infection.71-73 Several case reports described the recovery of ~lostridium sp. in purulent conjunctivitis associated with a foreign body, and anaerobic cocci have been reported in instances of corneal ulcers.” 3 We have recently observed the association of Bacteroides sp. as well as other anaerobic bacteria in conjunctivitis associated with sp. have been isolated from wearing contact lenses.74 Clostridium newborns with neonatal conjunctivitisU; however, no chlamydial cultures were taken. Although we have isolated additional Clostrichum sp. from inflamed conjunctiva in newborns, their exact role is yet to be determined. Anaerobes were sought in only one study of children with conjunctivitis.‘l The anaerobes (predominantly anaerobic cocci and Propionibacterium acnes) were isolated in 47 of 126 (37%) children with conjunctivitis and in 10 of 66 (15%) of controls (P < 0.05). These data are supported by several studies in adults.“, 73 Further studies in children are essential to evaluate the extent of the role of anaerobes in eye infection. Since anaerobes are also recovered from the normal noninflamed conjunctiva, inclusion of a control group is essential for any study. O&is Media-Recent developments in microbiological methodologies increased the awareness of the important role of anaerobic bacteria in ear infections. The predominance of anaerobic organisms in the oral flora, where they outnumber the aerobic bacteria in a ratio of ten to one,75 explains their involvement in otitis media (OM). Recognition of the role of anaerobic organisms in these infections has important implications on their management since some of these organisms are resistant to various antimicrobials. The recovery of anaerobic bacteria in OM depends on the utilization of proper and adequate techniques of collection, transport, and inoculation of the specimens.*6’*7 Many of the studies of acute, serious, and chronic otitis media did not employ methods for the recovery of these organisms.7J 76 However, whenever such techniques were utilized these organisms were isolated in substantial numbers.” 3 ACUTE OTITIS MEDIA.-Anaerobes, mainly gram-positive cocci, were recovered from 25% of 186 aspirates of acute OM, about a third of the time mixed with aerobic and facultative bacteria.37 Organisms were isolated in over 90% of the studied patients, a percentage larger than any other study, where methodologies for recovery of anaerobes were not utilized: However, since the external ear canal was not sterilized prior to aspiration, the possibility of contamination of specimen by the ear canal flora exists. A smaller study where strict sterilization of the ear canal was employed described the recovery of anaerobes in 3 of 28 infants.77 Since anaerobes recovered in acute OM are susceptible to penicilCur-r

Probl

Pediatr,

October

1987

677

lins and other antimicrobials commonly used to treat acute OM, no change in the recommended therapy is advocated. Studies are, however, advocated to further explore the role of anaerobes in acute OM. OTITIS MEDIA WITH EFFUSION .-In a study of OM with effusion that utilized methods for their cultivation, microorganisms were isolated in 23 of 57 (41%) of the patients studied.78 Anaerobes were present as the only isolate in 17% of the culture positive aspirates, and in an additional 26% they were present mixed with aerobes. The predominant anaerobes were gram-positive cocci and B. mekninogenicus group. The role of aerobic as well as anaerobic bacteria in the pathogenesis of OM with effusion is not yet clear; however, antimicrobial agents are often used in an attempt to clear the ear effusion of microorganisms that are generally recovered from about half of the cases. The presence of anaerobic as well as aerobic bacteria, many of which produce beta-lactamase, in serous ear aspirates raises the question of whether antibiotics also effective against these beta-lactamase-producing bacteria should be used. Controlled studies are needed to define the value of antimicrobial treatment in patients with OM with effusion and to clarify the role of bacteria in the pathogenesis of the inflammatory process.

CHRONICSUPPLJRATIW OTITISMEDIA.-The high number of negative cultures of chronic suppurative middle ear effusions in certain studies can be due to inappropriate methods for recovery of anaerobes.76 In one study,7 bacteria were seen on direct smears in 80% of middle ear effusions, while only 40% of the effusions yielded bacterial growth. Foul-smelling pus in the middle ear suggests the presence of anaerobic bacteria in many patients. Several studies reported the recovery of anaerobes in about half of patients with chronic suppurative OM38,7g--84(Table 5) and those with cholesteatomas.85, 86 The variability in the rate of recovery of anaerobes in these studies may be due to differences in geographical locations and in laboratory techniques. In several of the studies, the delays in cultivation were extensive and the length of incubation was inadequate for anaerobic bacteria. The predominant anaerobes were gram-positive cocci, Bacteroides sp., and Fusobacterium sp. Many of these organisms produced beta-lactamase and might have Fontributed to the failure of these patients to respond to penicillins. Support for the importance of anaerobes in chronic suppurative OM is provided by their recovery from most infectious complications of this infection. Anaerobes similar to those recovered in chronic suppurative OM were recovered from 23 (96%) of 24 specimens from patients with chronic mastoiditis.36 Anaerobes were recovered from most of the patients with intracranial abscesses complicating chronic suppurative OM.” 3,33 678

Curr

Probl

Pediatr,

October

1987

5.

of Recovq

Karma et Sugita et Aygagari Brook= Sweeney Constable Papastavms

et aI.” et al.= et aLW

al.‘9 al?’ et aL8’

AUTHOR (REFERENCE)

Frequency

TABLE

and

Aerobic

38/114 621760 68/1X 35/68 52/130 20/100 19/44

(33% 1 (8%) (59%) (51% ) (44% 1 (20%) (43%)

NUMBER OF CASES WERE AWEROBES WERE RECOVERED/ TOTAL NUMBER CASES (%I

of Anaerobic

15 38 33 31 7 9 10

Anaerobic cocci

Organisms

29 18 43 21 54 7 12

SP.

Bacteroides

Recovered

Suppurative

-

2 6 7 4 9

SP.

Fusobacterium

in Chronic

2 2 6 3 1 -

SD.

Clostridium

Otitis

32 8 22 15 33 29 2

Staphylococcus aureus

Media

16 12 29 33 25 15 20

SP.

Pseudomonas

37 31 32 31 86 34 4

Other gramnegative rods

Further evidence that anaerobes are important pathogens in chronic suppurative OM is provides by studies that showed efficacy of antimicrobials to which the organisms were susceptible in eradicating the infection in 52%85 to 89%“j of the patients. However, since these studies were done without a control group, further research is indicated to elucidate the benefit of medical and surgical management of this infection. A cholesteatoma that accompanies chronic suppurative OM can induce the absorption of bone. Various theories attempt to explain the factors involved in the process of expansion of a cholesteatoma and the collagen degradation that occurs in its vicinity.87 The production of organic acids by anaerobic bacteria may be involved in the destructive process.” Since a cholesteatoma associated with chronic suppurative OM contains bacteria similar to those recovered from chronically infected ears, the cholesteatoma may serve as a nidus of the chronic infection. Chronic Sinusitis.-Anaembes are frequently recovered from chronically inflamed sinuses. While they are generally isolated from only about 10% of patients with acute sinusitis, they can be recovered from up to two thirds of patients with chronic infections.2,3 The lowering of the oxygen pressure and increase in the bicarbonate contents in the sinuses occur as the infection persists, promoting the growth of anaembes as the infection becomes chronic. The predominant anaembes recovered are Peptostreptococcus sp. and Bacteroides sp .‘, 3 An average of almost three anaerobes per specimen was found in children with this infection.35 Sinus infections, when not treated promptly and properly, may spread via anastomosing veins or directly to nearby structures including the central nervous system. Orbital cellulitis is sometimes the first warning sign of extension of the infection.34 Intracranial complications include cavernous sinus thrombosis, retrograde meningitis, and epidural, subdural, and brain abscesses.33 The response of patients to antimicrobial therapy is often poor and unreliable even when appropriately given.34 This may be due to the poor penetration of the antimicrobial to the infected sinus, and inactivation of the antimicrobial by the pus. Controversies include how long antimicmbial therapy should be administered prior to surgical intervention. Surgical drainage is of paramount importance in these patients, especially if no clinical improvement has occurred, and should not be delayed more than 48 hours after an antimicrobial has been started. In many cases it should be an integral part of management. PharyngotonsiZZitis.-Group A beta-hemolytic streptococci, S. aureus, Streptococcus pneumoniae and viruses are traditionally associated with tonsillar and peritonsillar infections. However, anaerobes also have been isolated from the cores of tonsils of children with 580

Curr

Probl

Pediatr,

October

1987

recurrent GABHS62 and non-GABHS” tonsillitis and peritonsillar abscesses.43 Beta-lactamase-producing strains of B. fragilis, Fusobacterium sp. and S. aureus were isolated from the tonsils of 74% of children with GABHS recurrent tonsillitis6Z’g0-g3 (Table 6) and from 40% of children of non-GABHS tonsillitis.8s The possible role of anaerobes in the acute inflammatory process in the tonsils is supported by several clinical observations: the recovery of anaerobes as predominant pathogens in tonsil abscesses or the retropharyngeal area, in many cases without any aerobic bacteria43’94; their recovery as pathogens in well established anaerobic infections of the tonsils (Vincent’s anginaIg5; the increased recovery rate of encapsulated B. mehinogenicus in acutely inflamed tonsilss6; their isolation from the cores of recurrently inflamed nonGABHS tonsilssg; and the response to antibiotics in patients with non-GABHS tonsillitis.s7-‘01 Several recent studies where metronidazole was administered to patients with mononucleosis provided support of the role of anaerobes in tonsillitis.s7~ ” Metronidazole alleviated the clinical symptoms of tons&r hypertrophy and shortened the duration of fever. Metronidazole has no antimicrobial activity against aerobic bacteria and is effective only against anaerobes.lo2 A possible mechanism of its action could be suppression of the oral anaerobic flora that might have contributed to the inflammatory process induced by the Epstein-Barr viru~.~~‘~~ McDonald et al.” demonstrated a reduction in the severity of symptoms of adults with non-GABHS tonsillitis following the administration of erythromycin. Merenstein and Rogers”’ demonstrated definite improvement in the symptoms of patients with acute nonGABHS tonsillitis following penicillin therapy, as compared to placebo. Futto1o1 showed an earlier defervescence following penicillin therapy of children with non-GABHS tonsillitis as compared to patients with viral tonsillitis. All of these studies suggest that bacteria other than GABHS, including anaerobes, may be involved in acute tonsillitis. However, since no proof of that hypothesis is available, these observations have no practical implications at the present, and await further studies. TABLE

6.

Microbiology

of Excised

Tonsils

INVESTICATO&

Brook Reilly

Curr

Probl

Pediatr,

(268 Patients) NO.

%

BLPO

et al. U.SA., 1980g0 et al. U.K., 1981”

50 41

74 78

Tuner and Nerd Sweden, 1983” Chagollan et al. Mexico, 19849”

167 10

80

October

1987

73

581

Pulmonary Infections in Cystic Fibrosis.-Progressive pulmonary infections with S. aureus or Pseudomonas aeruginosa, and occasionally, other gram-negative organisms, play a major role in the morbidity and mortality of patients who suffer from cystic fibrosis (CF). Although anaerobes have been isolated from patients with aspiration pneumonia3’ and lung abscess,4o their possible role in CF was elusive. Two recent studies103’104 mvolving . small numbers of patients and utilizing culturing methods that bypass the normal oral flora described the recovery of mostly Bacteroides sp. with aerobic bacteria in patients with CF. Transtracheal aspiration was used in one study,‘03 where anaerobes were isolated from 3 of 6 patients, and thoracotomylo4 in the other, where anaerobes were isolated in 2 of 10 patients. The role of anaerobes in this infection is supported by the anecdotal successful use of antimicrobial agents effective against anaerobes (such as chloramphenicol) in CF patients. However, these data are preliminary and further prospective studies are needed to ascertain the exact role of these organisms in CF infections. ARE ANAEROBES IMPORTANT IN POLYMICROBIAL AEROBIC-ANAEROBIC INFECTIONS? The polymicrobial nature of anaerobic infection in the respiratory tract, abdomen, pelvis, and soft tissue is apparent in the majority of patients, where the number of isolates in an infectious site varies between two to five.‘, 3 The contributing role of anaerobic bacteria in these infections has often been questioned. An opinion voiced in the past was that it is sufficient to treat the aerobic component of the infectious flora to cure the infection.‘05 This simplistic attitude was based on the assumption that anaerobes are dependent on the aerobic and facultative component of the infection to lower the PO2 of their environment106 and to provide them with essential metabolic by-products.107 Therefore, elimination of the aerobic and facultative flora was thought to deprive the anaerobes of that support; hence, they could be eliminated by the host defenses. However, there are substantial clinical and laboratory data that disprove this hypothesis, and demonstrate the importance of anaerobes in polymicrobial infections. Some of these data will be discussed in this section. Another controversial issue in polymicrobial infections, including abscesses, is: Are all the organisms present in the abscess pathogenic, and should therapy be directed at the eradication of all or only a few of these? The following section will discuss these issues, and will provide data from research in animal models that shed light on these questions. These data provide evidence that support the conclusion that several aerobic and anaerobic organisms possessing a capsule are pathogenically important in such polymicrobial infections. The evi582

Cum

Probl

Pediatr,

October

1987

dence supporting the importance of anaerobic bacteria in mixed infections is based on clinical as well as animal studies, showing synergy between the aerobic and anaerobic bacteria, and the importance of encapsulation of anaerobic bacteria. Synergy Between Anaerobic and Aerobic or Facultative Bacteria.Polymicrobic infections are more pathogenic for experimental animals than those involving single organisms.10s Several studies document the synergistic effect of mixtures of aerobic and anaerobic bacteria in experimental infection. Altemeierlog demonstrated the pathogenicity of bacterial isolates recovered from peritoneal cultures after appendiceal rupture. Pure cultures of individual isolates were relatively innocuous when implanted subcutaneously in animals, but combinations of facultative and anaerobic strains manifested increased virulence. Similar observations were reported by Meleney et alno and Hite et al.ll’ Brook et all” evaluated the synergistic potentials between aerobic and anaerobic bacteria commonly recovered in clinical infections. Each bacterium was subcutaneously inoculated alone or mixed with another organism into mice, and synergistic effects were determined by observing abscess formation and animal mortality. The tested bacteria included encapsulated Bacteroides sp., Fusobacterium sp., and anaerobic cocci. Facultative and anaerobic bacteria included S. aureus, P. aeruginosa, E. coli, K. pneumoniae, and Proteus mirabilis. In many combinations the anaerobes significantly enhanced the virulence of each of the five aerobes. The most virulent combinations were between P. aeruginosa or S. aureus, and anaerobic cocci or Bacteroides sp. Enhancement of growth of aerobic and facultative bacteria was also apparent when they were coinoculated into mice and a subcutaneous abscess was formed. S. pyogenes, E. coli, S. aureus, K. pneumoniae, and P. aeruginosa were enhanced by B. flagiris, B. melaninogenicus,1138 I14 Peptostreptococci,“’ Fusobacterium sp.,l16 and Clos tridium sp .,l17 except C. d@iciZe. Although mutual enhancement of growth of both aerobic and anaerobic bacteria was noticed, the number of aerobic and facultative bacteria was greater than their anaerobic counterparts. Exceptions to the mutual enhancement were noticed in combinations between organisms that are generally not recovered together in mixed infections, such as Group D streptococci and B. melaninogenicus.114 The above observations suggest that the aerobic and fccultative bacteria benefit even more than the anaerobes from their symbiosis. The demonstration of the synergistic potentials of anaerobic bacteria commonly recovered in polymicrobial infections provides further support for their pathogenic role in these infections. Several hypotheses have been proposed to explain such microbial synergy. When this phenomenon occurs in mixtures of aerobic and anaerobic Curr

Probl

Pediatr,

October

1987

583

flora, it may be due to mutual protection from phagocytosis and intracellular killing,‘18 production of essential growth factors,lo7 and lowering of oxidation-reduction potentials in host tissues.lo6 In the latter situation, the resultant physical conditions are appropriate for replication and invasion by the anaerobic component nf the infection. Such environmental factors are known to be critical for anaerobic growth in vitro, and may be equally critical for experimental and human infections. Role of Encapsulated Bacteroides sp. and Peptostreptococcus sp. in Polymicrobial Infection.-An important virulence factor of Bacteroides sp. is the possession of a capsule. Several recent studies demonstrate the pathogenicity of encapsulated anaerobes and their ability to induce abscesses when injected alone in animals. Onderdonk et al.‘l’ correlated the virulence of B. j?agiZis strains with the presence of capsule, and Simon et a1.l” described decreased phagocytosis of the encapsulated B. fragilis. Capsular material from B. melaninogenicus also inhibits phagocytosis and phagocytic killing of other microorganisms in an in vitro system.lzl Tofte et al.,‘22 Jones and Gemmel, and Ingham et a1118 have shown that both phagocytic uptake and killing of facultative species were impaired by encapsulated Bacteroides. The presence of capsule in B. fragilis was shown to provide the organism with growth advantage in vivo over unencapsulated isolates.lz4 Furthermore, encapsulated strains survived better in vitro than unencapsulated variants when they were grown in an aerobic environment. Thus, the presence of a capsule apparently enables a strain of Bacteroides to resist exposure to oxygen as well as host defenses. Another recently described mechanism of protection was the inhibition of polymorphonuclear migration due to production of succinic acid by Bacteroides sp.lz5 We recently studied113, lz6,lz7 the importance of encapsulated Bacteroides sp. and Peptostreptococcus sp. by their ability to cause subcutaneous abscesses in mice. With few exceptions, encapsulated organisms were able to induce the subcutaneous abscesses when injected alone12”2 lz7 and possession of a capsule enabled these organisms to contribute more to the infectious process than their aerobic counterparts.113 Although unencapsulated Bacteroides sp. and Peptostreptococcus sp. did not induce abscesses, when isolates that contained only a small number of encapsulated organisms (< 1%) were inoculated with other abscess-forming viable or nonviable bacteria (“helpers”), the Bacteroides sp. survived in the abscess and became heavily encapsulated (Fig 1). Thereafter, these heavily encapsulated Bacteroides isolates were able to induce abscesses when injected alone. This phenomenon may explain how nonpathogenic organisms that are part of the normal mucous surface flora can 584

Curr

Probl

Pediatr,

October

1987

GaclcroidesmebninogcnkusGroup I

no-abscess

FIG 1. Summary genicus.

of the *“helper”

cycle of events leading to capsule = viable bacteria or formatinized

formation bacteria

by Bacteroides melaninoor capsular material.

become pathogens in chronic infections. Since the process of emergence of encapsulated strains takes 10 to 14 days, it may explain the higher frequency of recovery of anaerobic bacteria species in chronic infections, as compared to acute infections.‘, 3 The mechanism responsible for the phenomenon of emergence of a capsule is not yet known; future studies may show it to be due either to genetic transformation or to a process of selection. We have also evaluated the importance of encapsulated strains in clinical samplesl” by injecting them subcutaneously into mice alone or in all possible combinations with the other isolates and observed their ability to induce and/or survive in a subcutaneous abscess. Thirty-five isolates (30 anaerobes and five aerobes), 16 of which were encapsulated, were recovered from these abscesses. AU but one of the encapsulated organisms were able to cause abscesses by themselves, and were recovered from the abscesses when inoculated alone. Some strains that did not induce abscesses when injected alone survived only when they were injected with other organisms that were encapsulated. Two unencapsulated organisms were able to induce an abscess when injected together, suggesting that factors other than a capsule ,may also be important for abscess formation. The possession of a capsule by an organism was therefore found to be associated with greater virulence than the same organism’s unencapsulated counterpartsl” We found this to be true not only in Bacteroides sp., but also with anaerobic gram-positive cocci, Clostridium sp., and E. co/i. Possession of a capsule correlated with the Curr

Probl

Pediatr,

October

1987

686

ability of these organisms to cause an abscess by themselves, and might have allowed some of the other accompanying organisms to survive. Role of a Capsule of Bacteroides sp. and Anaerobic Cocci in Bacteremia.-Anaerobic bacteremia accounts for 5% to 15% of cases of bacteremia,’ and is especially prevalent in polymicrobial bacteremia, associated with abscesses.3z The role of possession of capsular material in the systemic spread of Bacteroides sp. and anaerobic and facultative gram-positive cocci (AFGPC) was investigated in mice following subcutaneous inoculation of encapsulated strains alone or in combination with aerobic or anaerobic facultative bacteria.“” Encapsulated anaerobes were isolated more frequently from infected animal blood, spleen, liver, and kidney than were nonencapsulated organisms. After inoculation with a single encapsulated anaerobic strain, encapsulated organisms were recovered in 163 of 420 (39%) animals, whereas nonencapsulated anaerobes were recovered in only 14 of 420 (3%) animals.12s Following inoculation of B. jkagilis mixed with aerobic or facultative flora, encapsulated B. fragilis was isolated more often and for longer periods of time than was the nonencapsulated strain. Furthermore, encapsulated B. fragilis was recovered more often after inoculation with other flora, than when it was inoculated alone. Therefore, encapsulated strains were found to be more virulent than their nonencapsulated strains. These data highlight the importance of encapsulated Bacteroides sp. and AFGPC in increasing the mortality associated with bacteremia and the spread to different organs. A similar pathogenic quality was observed in other bacterial species, such as S. pneurnoniae131 and H. injZuenzae,131 where the encapsulated strains showed greater ability for systemic spread. SigniJicance

of Anaerobic

Bacteria

in Mixed

Infection

With

Other

Flora.-Although anaerobic bacteria often are recovered mixed with other aerobic and facultative flora, their exact role in these infections and their relative contribution to the pathogenic process are unknown. The relative importance of the organisms present in an abscess caused by two species of bacteria (an aerobe and an anaerobe) and the effect of encapsulation on the relationship were determined by comparing abscess sizes in (1) mice treated with antibiotiks directed against one or both organisms, and (2) nontreated anim&.113,

116,117,127

As judged by selective antimicrobial therapy, the possession of a capsule in most mixed infections involving Bacteroides sp. generally made these organisms more important than their aerobic counterparts. In almost all instances, the aerobic counterparts in the infection were more important than nonencapsulated Bacteroides sp.‘13 586

Cur-r

Probl

Pediatr,

October

1987

Encapsulated members of the B. melaninogenicus group were almost always more important in mixed infections than their aerobic counterparts 6. pyogenes, S. pneumoniae, K pneumoniae, H. injhenzae, and S. aureus). Encapsulated B. flagilk group organisms were found to be more important than or as important as E. cob and enterococci and less important than S. aureus, S. pyogenes, and K. pneumoniae. In contrast to Bacteroides sp., encapsulated AFGPC were found more often to be less important than their aerobic counterparts.127 Clostridium sp. and Fusobacterium sp. were found to be less or equally important to enteric gram-negative rods116, ‘I7 Although Fusobacterium sp., AFGPC, and Clostridium sp. were generally equal to or less important than their aerobic counterparts, variations in the relationship existed. However, as determined by the abscess sizes, most of the anaerobic organisms enhanced mixed infection. CZinicaZ Signijicance of Capsule.-Detection of a capsule in a clinical isolate may suggest a pathogenic role of the organism in the infection. Two recent studies support the importance of encapsulated anaerobic organisms in respiratory infections.s6,132 The presence of encapsulated and abscess-forming organisms that belong to the B. melaninogenicus group was investigated in 25 children with acute tonsillitis and in a control group of 23 children without tonsillar inflammation.g6 Encapsulated organisms of the B. melaninogenicus group were found in 23 of 25 children with acute tonsillitis, compared to 5 of 23 controls (P < 0.0001). Subcutaneous inoculation into mice of the Bacteroides strains that were isolated from patients with tonsillitis produced abscess in 17 of 25 instances, compared with 9 of 23 controls (P < 0.045). These findings suggest a possible pathogenic role for the B. melaninogenicus group in acute tonsillar infection and suggest the importance of encapsulation in the pathogenesis of the infection. In another study,13’ the presence of encapsulated Bacteroides sp. and anaerobic gram-positive cocci was investigated in 182 patients with chronic orofacial infections and in the pharynx of 26 individuals without inflammation. One hundred seventy of the 216 (79%) isolates B. melaninogenicus and B. fkagilis groups, B. oralis, and anaerobic cocci were found to be encapsulated in patients with chronic infections, compared to only 34 of 96 (35%) controls (P < 0.001). Conclusions.-The recovery of a large number of encapsulated anaerobic organisms in patients with acute and chronic infections provides further support for the potential pathogenic role of these organisms. The presence of a capsule in a clinical isolate may add importance to the organism’s possible role as a pathogen in the infection. The demonstration of the importance of encapsulated organisms in mixed infection may justify directing therapy in such infections against these potential pathogens. Early and vigorous Curr

Probl

Pediatr,

October

1987

587

antimicrobial

therapy, directed at both aerobic and anaerobic bacin these mixed infections, may abort the infection before the emergence of encapsulated strains that contribute to the chronicity of the infection.

te&t

present

IS THE ROLE OF BETA-LACTAMASE-PRODUCING ANAEROBIC ORGANISMS IN THE FAILURE OF PENICILLIN THERAPY?

-T

Penicillins have been the agents of choice for the therapy of bacterial infections at various body sites. Within the last decade, an increased resistance to these drugs has been noticed. In addition to bacterial agents known to be penicillin-resistant (such as S. aureus and Enterobacteriaceae), other previously susceptible organisms showed increased resistance due to their ability to produce the enzyme beta-lactamase. These include aerobic bacteria such as Hemophilus influenzae,133 Branhamella catarrhalis,‘34 and anaerobic bacteria belonging to the family Bacteroidiaceae.61’135 The production of beta-lactamase is an important mechanism of virulence of Bacteroides sp. as well as of other aerobic and anaerobic bacteria. It is postulated that in this manner, these organisms not only are involved directly in the infection and protect themselves from the activity of penicillins, but may also shield penicillin-susceptible organisms from these drugs. This can occur when beta-lactamase is secreted into the infected tissues or abscess fluid in quantities sufficient to break the penicillins’ beta-lactam ring before it can kill the susceptible bacteria.5 The clinical importance of this phenomenon and its implication to patient management is still under investigation, and the therapeutic implication of this phenomenon is still not determined. The extent of distribution of beta-lactamase-producing bacteria (BLPB) in pediatric infection and several clinical and laboratory studies that provide support for this hypothesis will be described. Recovery of Beta-Lactamase-Producing Bacteria in Clinical Znfections.-Over the past decade, we have studied the aerobic and anaerobic microbiology of various infections in children.3 We also reported the recovery rate of BLPB in many of these infections. Betalactamase-producing bacteria were isolated in skin and subcutaneous tissue, upper respiratory tract, pulmonary and surgical infections’3”m’45 (Table 7). Beta-lactamase-producing bacteria were recovered in 288 (44%) of 648 patients with skin and soft tissue infections; 75% harbored aerobic BLPB and 36% had anaerobic BLPB (Tables 7 and 8). The infections where BLPB were most frequently recovered were vulvovaginal abscesses (90% of patients), perirectal and buttocks abscesses (79% 688

Curr

Probl

Pediatr,

October

1987

of patients), decubitus ulcers (64%), human bites (61%), and abscesses of the neck (58%). The predominant BLPB were S. aureus (68% of patients with BLPB) and the B. fragiks group (26% ). Beta-lactamase-producing bacteria were found in 262 (51% 1 of 514 patients with upper respiratory tract infections; 72% had aerobic BLPB and 57% had anaerobic BLPB (Tables 7 and 9). The infections in which these organisms were most frequently recovered were adenoiditis (85% of patients), tonsillitis in adults (82%) and children (74%), retropharyngeal abscess (71%), and chronic otitis media (57% 1. The predominant BLPB were S. aureus (49% of patients with BLPB), the B. melaninogenicus group (28% 1, and the B. fragilis group (20% ). Beta-lactamase-producing bacteria were isolated in 81 (59%) of 137 children with pulmonary infections; 75% had aerobic BLPB and 53% had anaerobic BLPB (Tables 7 and 10). The largest number of patients with BLPB was found in patients with cystic fibrosis (83% of patients), followed by pneumonia in intubated patients (78%), and lung abscesses (70% 1. The predominant BLPB were the B. fhagiZis group (36% of patients with BLPB), S. aureus (35%), B. melaninogenicus group (16%), P. aeruginosa (14%), K. pneumoniae (11%) and E. co/i (10% 1. Beta-lactamase-producing bacteria were recovered in 104 (92%) of 113 patients with surgical infections; 5% of the patients had aerobic BLPB and 98% had anaerobic BLPB (Tables 7 and 11). The most predominant BLPB was the B. fiagihs group. Beta-lactamase-producing bacteria were recovered in 16 (28%) of 57 children with miscellaneous infections (periapical and intracranial abscesses and anaerobic osteomyelitis). Twenty-five percent had aerobic BLPB and 80% had anaerobic BLPB (Tables 7 and 11). The rate of recovery of BLPB was not significantly different in these infections and BLPB were recovered most frequently in the B. melaninogenicus group (37% of patients), and the S. aureus and B. fragilis groups (25% each). The distribution of the different BLPB in infectious sites is similar to their distribution in the normal flora adjacent to the infected site.75 S. aureus that resides in the skin was mostly found in the infections of the skin and subcutaneous tissues. B. fiagilis that is part of the chronic flora was found in infections adjacent to that area, and B. melaninogenicus and B. ora&, ,which predominate in the oral cavity, were recovered in infections proximal to that area.‘j3 Zn Vivo and In Vitro Protection.-Several animal studies demonstrated the ability of bhe enzyme beta-lactamase to influence polymicrobial infections. Hackman and Wilkins146 showed that penicillin resistant strains of B. fragilis, B. melaninogenicus, and B. oralis protected a penicillin-sensitive Fusobacterium necrophorum from penicillin therapy in mice. Brook et al.,147 utilizing subcutaneous abscess Curr

Probl

Pediatr,

October

1987

689

TABLE

7.

Hecovery

v? Bacteria

of Beta-Lactamase-Pmducir

(BLPB)

NO. OF PATIENTS WITH BLPB/TOTAL NO. OF PATIENTS (%I WlTH BLPB

Skin/subcutaneous infection

(44%)

288/648

332

0%

90 of patients Upper tract

respiratory

151%) 262/514

o/10

O/6

344

68%

Pulmonary infections

8161

128/131

3%

49%

5%

(59%)

81/137

28/28

l/7

218

35%

1%

2%

104

% of patients

Other

12164

0% o/12

infection % of patients

Surgical

0%

o/5

infections

infections

(92%) 1041113 % of patients

(28%)

16157

113

o/3 0%

17

4/4 25%

O/l 0%

O/l 0%

356/361 47%

13/85 2%

2/25 0.3%

% of patients

(51%)

All infections

7441469

96 of patients

* = number

of BLPB/totaI

number

910

0%

8/67 1X

of strains.

models in mice, demonstrated protection from penicillin by B. fragilis and B. melaninogenicus. Clindamycin or the combination of penicillin and clavulanic acid (a beta-lactamase inhibitor), which are active against both GABHS and Bacteroides, were most effective in eradicating the infection. In vitro and in vivo studies have demonstrated the phenomenon of protection. A 200-fold increase in resistance of GABHS to penicillin was observed when it was inoculated with S. aureus.148 An increase in resistance was also noted when GABHS was grown with Haemophilus parainj7uenzae.14’ When mixed with cultures of B. fragdis, the resistance of GABHS to penicillin increased 8,500-fold.150 Beta-Lactamase in Clinical Infection-Several studies demonstrate the activity of the enzyme beta-lactamase in polymicrobial infections. de Louvois and Hurley151 demonstrated degradation of penicillin, ampicillin, and cephaloridine by purulent exudates obtained from four of 22 patients with abscesses. Studies by Matsuda and 590

Curr

Probl

Pediatr,

October

1987

11/46

3/15

16/31

19187

0%

7% 731191

4%

1%

3/26

6/26

331102

5/34

1%

2%

13%

2%

B/30

9/40

11137

10%

11%

5/61 4%

l/11

6%

o/11

14%

Z/5 2% o/2

1%

0%

Z/3

75175

S/63

1%

0.6%

26%

3%

19145

Z/14

52152

3198

7%

1%

20%

1%

29/29

o/11

36%

0%

O/26

102/102

0%

98%

S/23 5%

28% 13/59 16%

6/24 37% 271163 4%

19/92 3%

60/170 8%

7/52 1%

111/387 15%

z/9

O/l 0%

l/9 1%

Z/7

4/4

12%

25%

23162 3%

S/26 1%

262/262 35%

l/l0 6% 17/205 2%

demonstrated possible beta-lactamase activity in emTomioka”’ pyema fluid. Most infections were polymicrobial and involved both K. pneumoniae and P. aeruginosa. O’Keefe et a1.153demonstrated inactivation of penicillin G in an experimental B. fragiZis infection model in the rabbit peritoneum. The presence of beta-lactamase in clinical specimens was reported by several investigators. Bryant et al.154 studied the beta-lactamase activity in samples of pus obtained from 12 patients with polymicrobial intra-abdominal abscesses or polymicrobial empyema. Using the chromogenic cephalosporin nitrocefin, these investigators were able to show strong enzyme activity in four of the 11 abscess who studied cerebrospinal specimens using the fluids. Boughton, nitrocefin reagent, was able to detect enzyme activity in all of the five specimens that contained beta-lactamase-producing H. influenzae, in none of the 33 specimens that contained non-beta-lactamase-producing H. injluenzae, and in none of the 234 sterile specimens. Curr

Probl

Pediatr.

October

1987

591

TABLE

8.

Beta-Lactamase-Pmduci

w

Anaerobic

Bacteria

in Skin

and

-

Soft Tissue

Infections

2% e 22 8F.Y QO

Head abscess Neck abscess Extremities and trunk Perirectal buttocks Pilonidal (cyst) Vulvovaginal Cervical lymph adenitis Burns Pamnychia Decubitus ulcer Bite, animal Bite, human Omphalitis Total * = numberofBLPB/totd

10131 (32%) Xl/36 (58%) 56/90 (62%) 34/43 (79%) 10123 (43%) 4/5 (90%) 15/45 (33%) 31/180 (17%) 15/33 (45%) 64/100 (64%) 5/21 (24%) 11/18 (61%) 12123 (53%) 288/648 (44%) number

20 34 77 32 13 5 35 458 51 125 37 44 47 968

42 18 66 104 63 18 31 132 67 90 22 53 22 728

3/7* l/3 l/6 l/14 2/10 3/14 2111 Z/7 Z/6 o/3 2/6 19/87

012

s8 $2 ij ‘B ag 2/z 7/7 22/22 lO/lO 4/4 5/5 l/l 16/16

II4

l/3

l/l 717 7975

219

136 136 28 28,137 138 28 139 29 58 31 10 10 45

2/12 4119 l/14 l/3 o/2 o/3

O/8 o/2 8163

of strains

Yolken and Hughesls6 have recently described ac direct method for detecting beta-lactamase in body fluids which requires radioisotopes and radioisotope-measuring equipment. Brook143 studied beta-lactamase activity in pus obtained from 109 abscesses. Beta-lactamaseproducing organisms were recovered in 84 (77%) specimens. These included all 28 isolates of B. fragilis, 18 of 30 B. mehinogenicus, 42 of 43 S. aureus and 11 of 14 E. co/i. Beta-lactamase activity was detected in 46 (55%) of the purulent specimens when using the chromogenic, cephalosporin nitrocefin methodology. The recovery of penicillin-susceptible bacteria mixed with BLPB in patients who have received penicillin or cephalosporin therapy and failed to respond to therapy, suggests the ability of BLPB to protect a penicillin-susceptible or a cephalosporin-susceptible organism from the activity of those drugs.

Clinical Studies.-The selection of BLPB following antimicrobial therapy may account for many of the cases of clinical failures after penicillin therapy. A recent report described five adults with clinical failures after penicillin therapy associated with the isolation of anaerobic BLPB .ls7 In a study of 185 children with orofacial and respiratory infections who failed to respond to penicillin, BLPB were recovered in 75 (40%).4 The predominant BLPB were S. aureus, B. melaninogenicus gr., B. fi-agilis gr., and B. oralis. 592

Cut-r Probl

Pediatr,

October

1987

Bacteroides

6acteroide.s

Bacteroides ruminicola

Bacteroides

sp.

i

25 roTAL NUMBER DF AEROBES s 4 l-0T.u NUMBER 93 3FANAEROBES g.

1EFERENCE UUMBER

Bacteroides

Bacteroides fi-agilis

Bacteroides ruminicola

Bacteroides oralis

Bacteroides melaninogenicus

sp.

RX’AL NUMBER “F ANAEROBES

TOTAL NUMBER OF AEROBES

NUMBER PATENTS W,TH BLPB/ TOTAb NUMBER PATIENTS ( % ) WlTH BLPB

TABLE

11.

Beta-Lactamase-Producin

Intra-abdominal abscess Peritonitis (appendicitis) Biliary tract Total

4naerobic

Bacteria

in

nd

Miscellaneous

Infer

----r-l---i717

95/100 Z/6

1041113

(100%) (95%)

(33%)

192%)

11 144

6/6’

94/94

31

5123

2/2

186

102/102

5/23

of GABHS in the pharynx despite treatment with intramuscular penicillin in 19% of the patients after the first course of therapy and in 43% of the remainder of the patients after retreatment. Various theories have been offered to explain this penicillin failure. One theory is that repeated penicillin administration results in a shift in the oral microflora with selection of beta-lactamase-producing strains of Hemophilus sp., S. aureus, B. catarrhalis and Bacteraides sp .go,148,14g,ls7 It is possible that these BLPB can protect the GABHS from penicillin by inactivation of the antibiotic. Clinical evidence supporting the ability of BLPB to protect a penicillin-susceptible pathogen was first suggested in 1963.‘48 Knudsin and Miller15’ demonstrated a significantly higher carrier rate of penicillin-resistance S. aureus in patients with penicillin treatment failure than in patients with treatment success. In contrast, Quie et al.,160 who did not use methods for detection of anaerobic BLPB, found no correlation between the presence of penicillinase producing S. aureus before therapy or at follow-up in penicillin treatment failures or success. Recent studies suggested that H. parainfluenzae14g and B. failure. The role of catarrhalis’34 may also have a role in penicillin anaerobic BLPB in persistence of GABHS was suggested by Brook et al.” who studied core tonsillar cultures recovered from 50 children suffering from recurrent tonsillitis. One or two strains of aerobic and/ or anaerobic BLPB were recovered in 74% of the tonsils. The anaerobic BLPB included strains of B. j?agiZis and B. melaninogenicus groups and B. oralis, while the aerobic bacteria were S. aureus, He594

Curr

Probl

Pediatr,

October

1987

mophilus sp., and B. catarrhalis.gof’61 Assays of the free enzyme in the tissues demonstrated its presence in 33 of 39 (85%) tonsils that harbored BLPB, while the enzyme was not detected in any of the 11 tonsils without BLPB.16’ This observation was confirmed by Reilly et algl who found penicillin resistance in 78% of Bacteroides isolated from tonsils, Chagollan et al.,g3 who isolated BLPB in eight of ten patients, and Tuner and Nerd,” who recovered aerobic and anaerobic BLPB in 122 of 167 (73%) of their patients (see Table 6). B. ruminicola accounted for 98 of 202 beta-lactamase-producing Bacteroides sp. recovered by Tuner and Nerd,” who have also recovered Fusobacterium nucleatum strains that produce beta-lactamase from infected tonsils.“5 Brook and Gober163 and Tuner and Nord*64 have demonstrated the rapid emergence of BLPB following penicillin therapy. Brook and Gober isolated BLPB in 3 of 21 (41%) children prior to penicillin therapy, and in 10 of 21 (48%) following one course of penicillin. The organisms were members of B. meZaninogenicus group, S. aureus, B. catarrhalis and Haemophilus injkenzae. Surveillance of 26 patients treated with penicillins165 showed continuous harboring of BLPB in 35% of the patients 40 to 45 days after completion of therapy and in 27%, 8.5 to 90 days after therapy. These organisms were also isolated from household contacts of children repeatedly treated with penicillin, suggesting their possible transfer within a family.163 Tuner and Nord,164 who treated 10 healthy volunteers for 10 days, observed a significant increase in the number of beta-lactamase-producing Bacteroides sp. (from 8 of 35, to 21 of 30), and Fusobacterium nucleatum (from 1 of 10, to 3 of 7). Beta-lactamase activity in saliva was observed in all volunteers and paralleled the increase of BLPB. A recent study demonstrated the association between the presence of BLPB even prior to therapy and the outcome of lo-day oral penicillin therapy.166 Of 98 children with acute GABHS tonsillitis, 36 failed to respond to therapy (Table 12). Prior to therapy, 17 isolates of BLPB were detected in 16 (26%) of those cured, and following therapy, 30 such organisms were recovered in 19 (30%) of these children. In contrast, prior to therapy 40 BLPB were recovered from 25 (69%) of the children who failed to respond to therapy, and following therapy, 62 such organisms were found in 31 (86%) of the children in that group. Ross et al.‘67 showed that high levels of beta-lactamase in saliva may reflect colonization with large numbers of BLPB. These investigators demonstrated that patients with recurent GABHS tonsillitis had detectable amounts of beta-lactamase in their saliva compared to patients with uncomplicated courses of tonsillitis. The data presented so far indicate the increasing role of BLPB in mixed infections. They also demonstrate the rapidity in which BLPB can appear in patients and spread to other household members. Curr

Probl

Pediatr,

October

1987

696

TABLE

12.

Beta-Lactamase-Producing

Isolates

Recovered

From

Children*

GROUP 1

GROUP II

ISOL4TES REcovl?.REIJ IN 21

P!3NICILLIN TKEATED CHILoREN OROANISMS

BEFORE THEFLAPY

Bacteroides Staphylococcus aureus Total number children ‘Modified to eradicate

sp.

AFTER THERAPY

ISOLATES RECOVERED IN 18 NON-TREATEO CHILImEN BASELINE

FOLLOW-UP

23 (0)

26 (8)

23 (8)

20 (0)

2 (21

4 (3)

1 (1)

2 (11

2 (11%)

116%)

of

3 (14%)

10 (48%)

fmm Brook I: Role of beta-lactamase-producing bacteria Group A streptococci, Pediarr Infect Dis 1985; 4:491.

in penicillin

failure

Anaerobes, therefore, have two mechanisms of virulence: direct and indirect. The direct mechanism is through possession of a capsule and production of enzymes and toxins detrimental to the host; the indirect mechanism is through the production of beta-lactamase. Implications

of Direct

and

Indirect

Pathogenic@

on

Therapy.-

The importance of anaerobic bacteria in mixed infection highlights the need to direct antimicrobial therapy against them. Animal and patient studies have demonstrated that in mixed infections, unless therapy is directed against both aerobic and anaerobic bacteria, the untreated organisms will survive.16a172 Weinstein et al.‘“8 demonstrated in an intra-abdominal abscess model in rats, that combined therapy of clindamycin and gentamicin was needed to prevent mortality due to E. coli sepsis and abscesses due to B. frasilis. Thadepalli et al.16’ showed that in patients with intra-abdominal trauma, clindamycin and gentamicin were superior to cephalothin and kanamycin in preventing septic complications. This principle of double coverage against aerobes and anaerobes has since been proved to be the golden standard of therapy in numerous studies,170-*7Z utilizing combination therapy (clindamycin, metronidazole or cefoxitin plus an aminoglycoside) and more recently, single agent therapy with cefoxitin’73 or imipenem.174 A similar approach was found essential in the management of pelvic inflammatory disease, where mixed aerobic-anaerobic flora are recovered from the majority of cases.64 The studies showing the process of encapsulation of Bacteroides sp., which are assisted by aerobic or facultative “helpers,” provide insight into the synergistic relationship between these organisms. Since even killed “helpers” provided protection for the unencapsulated anaerobic bacteria, antimicrobial therapy should be directed 696

Cur-r Probl

Pediatr,

October

1987

against both the aerobic and anaerobic components of mixed infection. The presence of BLPB in mixed infection warrants administration of drugs that will be effective in eradication of BLPB as well as the other pathogens. The high failure rate of penicillin therapy associated with the recovery of BLPB in a growing number of cases of mixed aerobic-anaerobic infections underscores the importance of this therapeutic appmach.4,‘57 An infection where this therapeutic approach has been successful is recurent tonsillitis. Antimicrobial agents active against BLPB as well as GABHS have been shown to be effective in the eradication of this infection. Several studies demonstrated the efficacy of lincomycin175-178 and clindamycin16” 17s-182and the combination of penicillin and rifampin183J ‘~4 over penicillin (Table 13). The most recent of these studiesl’l was a double-blind study comparing penicillin to erythmmycin and clindamycin. Erythmmycin was chosen for its effectiveness against S. aureus and Group A streptococci; clindamycin was chosen for its effectiveness against S. aureus, Bacteroides and Group A streptococci. With penicillin therapy, there were only two cures out of 15, with erythromycin, six of 15, and with clindamycin 14 of the 15 patients recovered. Four patients who received penicillin and two who received erythmmycin required a tonsillectomy. No tonsillectomies were required in the clindamycin group. Another recent study compared the efficacy of clindamycin to penicillin in the therapy of lung abscesses.“’ Clindamycin was superior to penicillin in terms of initial response rates, time of defervescence, and clearing of fetid sputum. The superiority of clindamycin over penicillin was postulated to be due to its ability to eradicate the beta-lactamase-producing Bacteroides sp. present in lung abscesses. Other drugs that may be effective in the therapy of this condition TABLE Studies

13. of Therapy

of Acute

Group

PENIClLLlNS

INVESTIGATORS

Breese et al. 1966 and 19691”“‘6 Randolph and DeHaan, 1969l” Howie and Plousard, 1971’.78 Randolph et al. 19701” Stillerman et al., 1973l” Massell 1979 (pmphylaxis)‘81 Chaudhary et al., 19851s3 Tanz et al., 1985 (carriers)184

Cur-r

Pmbl

Pediatr,

A Streptococcal

October

1987

38/131 37/267 32/80 16/72 9/51 26/102 11/39 3/10

(29%) (14%) (40% ) (22%) (18%) (25% ) (28% 1 (30%)

Pharyngitis

OTHER DRUGS

Lincomycin 17/131 (13% 1 Lincomycin 20/2.58 (8% 1 Lincomycin lo/76 (13%) Clindamycin 4/.56 (7% 1 Clindamycin 5kS.Z (10% 1 Clindamycin 12/100 (12% 1 Penicillin and &unpin O/60 (0%) Penicillin and rifampin l/14 (7%)

597

are the combination of a beta-lactamase inhibitor-clavulanic acid and amoxicillin, and the combination of metronidazole and erythromycin or spiramycin. In infections that involve gram-negative organisms, some of which can also produce beta-lactamases, therapy should be directed against these bacteria. This can be achieved by the addition of agents such as aminoglycosides or third-generation cephalosporins, or the administration of agents with a wider spectrum of activity such as primaxim. Summary-Controversies regarding the role of anaerobic bacteria in pediatric infection arise from uncertainty regarding their importance in polymicrobial infections. However, studies have demonstrated their potential synergy with aerobic organisms, due to several pathogenic mechanisms. These mechanisms include the production of a capsule, which protects organisms from phagocytosis, and the production of beta-lactamase, which inactivates penicillins and first generation cephalosporins. Both mechanisms enhance infection and induce complications. Data, thus far, indicate that there are therapeutic modalities that can be utilized to prevent these occurrences and eliminate many mixed infections. CONTROVERSIES INFECTIONS

IN MANAGEMENT

OF

ANAEROBIC

The principles of management of anaerobic infections include neutralization of toxins produced by anaerobes, preventing the local proliferation of anaerobes by changing the environment, and hampering their spread into healthy tissues. Toxin neutralization by specific antitoxins may be employed especially in infections caused by Clostridium sp. (tetanus and botulism). Control of the environment is achieved by debridement, drainage of pus, improvement of circulation, alleviation of obstruction, and increase of tissue oxygenation. The primary role of antimicrobials is limiting the local and systemic spread of the organism. The controversies that arise in the management of anaerobic infection center mainly on these issues: (1) the role of hyperbaric oxygen (HBO) therapy; (2) the choice between surgical or medical therapy; and (3) which antimicrobial to use for the management of the infection, and for how long. What Is the Role of Hyperbaric Oxygen?-There is controversy whether HBO should be used in infection of spore-forming grampositive anaerobic rods. There are several uncontrolled reports that demonstrate efficacy in individual cases.ls6 However, since no wellcontrolled studies are available, the use of HBO is unproven. There is no contraindication to its use in conjunction with other therapeu698

Curr

Probl

Pediatr,

October

1987

tic measures except when it may delay the execution of other essential procedures. Topical application of oxygen-releasing compounds may be useful as an adjunct to other procedures. Surgical vs. Medical Therapy.-In many cases, surgical therapy is the most important and sometimes the only form of treatment required, whereas in other cases it is an important adjunct to a medical approach. Surgery is important in draining abscesses, debriding necrotic tissues, decompressing closed space infection, and relieving obstruction. Drainage of pleuropulmonary abscesses except empyema is usually contraindicated, as the abscesses may spread to other lung tissues during the procedure. Percutaneous or catheter drainage of intra-abdominal abscess, under ultrasound or computed tomography guidance, is employed as a substitute for surgery with increasing frequency. Drainage of an intracranial abscess is generally mandatory. The urgency in performing the surgery depends on whether intracranial pressure has increased. However, in the early stages of the disease where only cerebritis exists, and a capsule around the abscess has not yet been formed, antimicrobial therapy may be curative. Antimicrobial therapy only may be indicated in patients with multiple abscesses. This approach has been found to be therapeutic by itself,ls7 and may be considered in high-risk patients as well as those with multiple abscesses. The Choice ofAntimicrobial Agents.-Appropriate management of mixed aerobic and anaerobic infections requires the administration of antimicrobials that are effective against both aerobic and anaerobic components of the infection,l13 in addition to surgical correction and drainage of pus. When such therapy is not given, the infection may persist and more serious complications may occur.16gS17o A number of factors should be considered in the choice of appropriate antimicrobial agents. They should be effective against all target organism(s), induce little or no resistance, achieve sufficient levels in the infected site, produce minimal toxicity, and have maximum stability and longevity. Antimicrobials often fail to cure the infection. They may fail because of bacterial resistance, insufficient tissue levels, incompatible drug interaction, or abscess development. The environment of an abscess is detrimental for many antimicrobials. The abscess capsule interferes with the penetration of antimicrobial agents, and the low pH and the presence of binding proteins or inactivating enzymes (beta-lactamase) may impair the activity of many antimicrobials. The low pH and the anaerobic environment within the abscess are especially deleterious toward the aminoglycosides.‘88 It should also be remembered that an acidic environment, high osmolarity, and presCum

Probl

Pediatr,

October

1987

599

ence of an anaerobic environment can develop in an infection site without the presence of an abscess.lo5 When choosing antimicrobials for the therapy of mixed infections, their aerobic and anaerobic antibacterial spectrum and their availability in oral or parenteral form should be considered (Table 14). Some antimicrobials have a limited range of activity. Metronidazole is active only against anaerobes, and therefore cannot be administered as a single agent for the therapy of mixed infections. Others (i.e., cefoxitin, imipenem) have a wider spectrum of activity against Enterobacteriaceae and anaerobes. The selection of antimicrobial agents is simplified when a reliable culture result is available. However, this may be particularly difficult in anaerobic infections, because of the problems in obtaining appropriate specimens. For this reason, many patients are treated empirically on the basis of suspected, rather than identified, pathogens. Fortunately, the types of anaerobes involved in many anaerobic infections and their antimicrobial susceptibility patterns tend to be predictable.24’ lBg However, some anaerobic bacteria have become resistant to antimicrobial agents, and many can develop resistance while a patient is receiving therapy.163’ 164 The susceptibility of the B.fragilis group, the most commonly recovered group of anaerobes, to the commonly used antimicrobial drugs was studied systemically over the past several years by Tally et aLz4 who collected several hundred strains each year from several medical centers across the United States. These surveys have shown TABLE

14.

Antimicrobial

Agents

Effective

Against

Mixed

Infection*

ANABROBIC BACTERIA ANTIMICROFOAL AGENT

BETA-LKTAMASE-PRODUCING BAC’IXROIDES

Penicillint Chloramphenicolt Cephalothin Cefoxitin Moxalactam Imipenem Clindamycint Carbenicillin Amoxicillin plus Clavulanic acid-t Ticarcillin plus Clavulanic acid Metronidazolet * = Degrees of actitity: 0 t = Available also in oral 600

AEROBIC BACTERIA OTHER Ah!AEROBES

GRAM-POSITIVE COCCI

ENTERICS

+++

+++ +++ + +++ +++ +++ +++ +++ +++

+ + ++ + + +++ +++ + ++

+++

+++

++

++

+++

+++

0

0

0 +++ 0 +++ ++ +++ +++ +

0 + +/++ +++ +++ D ++ ++

to + + + form. Curr

Probl

Pediatr,

October

1987

no resistant strains to chloramphenicol and metronidazole; resistance to other agents varies. Resistance differs among the contributing centers, and generally increases with the extensive use of certain antimicrobial agents (penicillins, cephalosporins, and clindamycin). Aside from susceptibility patterns, other factors influencing the choice of antimicrobial therapy include the pharmacologic characteristics of the various drugs, their toxicity, their effect on the normal flora, and their bactericidal activity.‘j3 When anaerobic bacteria are recovered mixed with aerobic organisms, a common occurrence,” 3 selecting proper therapy is more complicated. In the treatment of mixed infection, the chosen antimicrobial agents should adequately cover most of the pathogens. Some single broad-spectrum agents possess such qualities, but for certain organisms, additional agents should be added to the therapeutic regimen. Antimicrobial therapy usually should be given for prolonged periods for anaerobic infections because of their tendency to relapse. Penicillins.-Penicillin G is the drug of choice when the infecting strains are susceptible. This includes the vast majority of anaerobic strains other than those belonging to the B. fragilis group.“’ Other strains that may show resistance to penicillins are growing numbers of Bacteroides sp. (e.g., B. melaninogenicus, B. oralis, B. bivius, B. disiens), Clostridium sp., Fusobacterium sp., and microaerophilic streptococci. Some of these strains show minimal inhibitory concentration (MIC) in dosages of 8 p. to 32 p per ml of penicillin G. In these instances, administration of very high doses of penicillin G may eradicate the infection. Ampicillin, amoxicillin, and penicillin generally are equally active, but the semisynthetic penicillins are less active than the parent compound. Methicillin, nafcillin, and the isoxazolyl penicillins have unpredictable activity and frequently are inferior to penicillin G against anaerobes. Clauvulanic acid is a new beta-lactamase inhibitor that resembles the nucleus of penicillin but differs in several ways. Clavulanic acid irreversibly inhibits beta-lactamase enzymes produced by some Enterobacteriaceae sp., staphylococci and beta-lactamase-producing Bacteroides sp. CB.fiagilis group and strains of B. melaninogenicus and B. oraZis).1go8lgl When used in conjunction with a beta-lactam antibiotic, it is effective in treating infections caused by beta-lactamase-producing bacteria. Two available agents are a combination of clavulanate and amoxicillin, and a combination of clavulanate and ticarcillin. The usefulhess of clavulanic acid in the treatment of human infections is currently being evaluated in clinical trials. Clavulanic acid and other beta-lactamase inhibitors may be effective adjuncts to penicillins in the treatment of resistant organisms. The semisynthetic penicillins (carbenicillin, ticarcillin, piperacillin, Cum

Probl

Pediatr,

October

1987

601

and mezlocillin) are generally administered in large quantities to achieve high serum concentration. These drugs are effective against Enterobacteriaceae and have good activity against most anaerobes in these concentrations. However, they are not absolutely resistant to beta-lactamase produced by Bacteroides sp.,lBg and up to 30% of the B.fiagiZis group are resistant to these agents. CephaZosporins.-The activity of cephalosporins varies against the beta-lactamase-producing Bacteroides sp. The antimicrobial spectrum of the first-generation cephalosporins against anaerobes is similar to penicillin G, although on a weight basis they are less active. Most strains of the B.fragiZis group and many of the B. melaninogenicus group are resistant to these agents by virtue of cephalosporinase production.” lBg Cefoxitin, a second-generation cephalosporin, is relatively resistant to this enzyme and is therefore the most effective cephalosporin against the B.fragiZis group. Cefoxitin is active in vitro against at least 95% of B. fragiZis strains at a level of 32 pg per ml. Cefoxitin is relatively inactive against most species of Clostridium (including C. d@ciZe); C. pe@ingens is an exception.2’18g Clinical experience with cefoxitin in anaerobic infections show it to be effective in eradication of these infectionsl” This drug is often used for prophylaxis for surgery involving body sites with mucous membranes because of its activity against enteric gram-negative rods as well. The third-generation cephalosporins, except moxalactam, are not as active against B.@agiZis as cefoxitin. However, these agents have improved activity against Enterobacteriaceae. The third-generation drugs do not possess any advantage over cefoxitin in surgical prophylaxis. There is some confusion about the use of cephalosporins to treat the B. fragiZis group. B. j?agiZis group is the most important anaerobic pathogen recovered from intra-abdominal infections. The B. @agiZis group is composed of several Bacteroides species that recently were promoted to a genus level. Among the B. fragiilis group, B. j?agiZis accounts for 40% to 54% of the Bacteroides isolates recovered from intra-abdominal infections.17’30 However, another important pathogen that belongs to the B.fragiZis group is Bacteroides thetatiotaomicron, which accounts for 13% to 23% of the isolates. Other members of the B.fragiZis group account for 33% to 37% (Table 15). The antimicrobial susceptibility of members of the B. fragilis group varies, especially to the secondand third-generation cephalosporins. While B. fiagilis is generally the most susceptible, B. thetatiotaomicron and B. diastasonis generally are more resistant.57 ChZoramphenicoZ.-Although it is a bacteriostatic agent, chloramphenicol is one of the antimicrobial agents that is most active against anaerobes, and resistance to this drug is rare.2’18s It has been used for over 25 years to treat anaerobic infections, and is regarded 602

Curr

Probl

Pediatr,

October

1987

TABLE

15.

Incidence of Bacteroides ji-agilis Group in Intra-abdominal Infection in Adults (185 isolates)* and Children (100 isolate&t INCIDENCE

Bacteroidk

f?a.cilis

Gr.

Bacteroides Bacteroides Bacteroides Bacteroides Bacteroides Bacteroides Bacteroides Bacteroides

ji-agilis thetatiotaomicron diastaaonis vulgatus ovatus uniformis splanchnicus eggerthii

(% )

ADULTS

CHILDREN

40 23

54 13

37

33

*Data from Sutter VL, Citron DM, Edelstein MAC, et al: Anaerobic Bacteriology Manual, ed 4. Belmont, Cala Star Publishing co, 198.5. tData fmm Brook I: Bacterial studies of peritoneal cavity and postoperative surgical wound drainage following perforated appendix in children. Ann Surg 1980; 46524.

as a good choice for treating serious anaerobic infections when the nature and susceptibility of the infecting organisms are unknown, and for treating infections of the central nervous system because of its good penetration through the blood-brain barrier. Despite chloramphenicol’s effectiveness, its toxicity must be borne in mind. This includes the low risk of aplastic anemia, dose-dependent leukopenia and gray baby syndrome in the newborn. CZindamycin.-Clindamycin has a broad range of activity against anaerobic organisms, and it has proved its efficacy in clinical trials.2, lSS Approximately 95% of the anaerobic bacteria recovered from clinical specimens are susceptible to easily achieved levels of clindamycin. B. fragih is generally sensitive to clindamycin concentrations as low as 3 kg per ml. There are, however, reports of resistant strains associated with clinical infections,z4 although these are not common. Because of its effectiveness against anaerobes,ls3 clindamycin is frequently used in combination with aminoglycosides for the treatment of mixed aerobic-anaerobic infections of the abdominal cavity and for obstetric infections.ls3 Clindamycin does not cross the blood-brain barrier efficiently and should not be adminigtered in central nervous system infections. The primary manifestation of clindamycin toxicity is colitis.66 Colitis has also been associated with a number of other antimicrobial agents, such as ampicillin and all the cephalosporins, and has been described in seriously ill patients in the absence of previous antimicrobial therapy. Fortunately, the occurrence of colitis in pediatric Curr

Probl

Pediatr,

October

1987

603

patients is rare.ls4 Clinical studies using clindamycin in pediatric patients have shown it to be effective in the treatment of intra-abdominal infections,ls5 aspiration pneumonia,1s6 and chronic otitis media.= MetronicZazoZe.-Metronidazole shows excellent bactericidal activity against most obligate anaerobic bacteria, such as B. fragilis and other Bacteroides sp., Fusobacterium and CZo.stridium.18s Occasionally, strains of anaerobic gram-positive cocci and nonsporulating bacilli are resistant to metronidazole. Microaerophilic streptococci, P. acnes, and Actinomyces sp. are almost uniformly resistant. Aerobic and facultative anaerobes, such as coliforms, are usually highly resistant. Over 90% of obligate anaerobes are susceptible to metronidazole concentrations of less than 2 kg per ml. Vast clinical experience in adults”’ and limited experience in children1s7 indicate that metronidazole is effective in the treatment of infections caused by anaerobic bacteria, including intra-abdominal sepsis, infections of the female genital tract and central nervous system infections. However, this drug does not seem to be as effective in therapy of anaerobic gram-positive pulmonary infections. Because of its lack of activity against aerobic bacteria, antimicrobial agents effective against aerobic bacteria should be administered whenever these organisms are also present. The use of metronidazole seems advantageous in central nervous system infections because of its excellent penetration into the central nervous system.“’ Until the use of metronidazole is approved by the United States Food and Drug Administration for children, this drug should be used only in seriously ill pediatric patients. TetracycZines.-Tetracycline has limited usefulness in anaerobic infections because of the development of resistance by all types of anaerobes. Only about 45% of all B.fiagiZis strains are susceptible to this drug.” Is9 The tetracycline analogs, doxycycline and minocycline, are more active than the parent compound. The use of tetracyclines is not recommended for children under 8 years of age because of adverse effects on teeth. Vancomycin.-Vancomycin is effective against all gram-positive anaerobes but is inactive against gram-negative anaerobes.2’18s Little clinical experience has been gained in the use of this agent for the treatment of anaerobic bacteria. Zmipenem.-Imipenem, a thienamycin, is a new beta-lactam antibiotic that is effective against a wide variety of aerobic and anaerobic gram-positive and gram-negative organisms.174 It also possesses excellent activity against beta-lactamase-producing Bacteroides sp. This new drug may prove to be an effective single agent for the therapy of mixed aerobic-anaerobic infections. 604

Curr

ProbI

Pediatr,

October

1987

Suggested choices of the different antimicrobials is summarized in Table 16. Prophylactic therapy prior to surgery is generally administered when the area of surgery is expected to be contaminated by the normal mucous membrane at the operated site. Cefazolin, a firstgeneration cephalosporin, is generally effective in surgical prophylaxis in sites distant from the oral or rectal areas. Cefoxitin is the drug of choice in surgical prophylaxis in procedures that evolve the mucous surfaces (oral, rectal or vulvovaginal) because of its ability to cover the aerobic and anaerobic flora at the operation site that evolve in most mucous surfaces. The parenteral antimicrobials that can be used in most infectious sites are clindamycin, metronidazole, chloramphenicol, cefoxitin, ticarcillin plus, clavulanic acid, and imipenem. Aminoglycosides are generally added to clindamycin, metronidazole, and occasionally cefoxitin, when treating intra-abdominal infection to provide coverage for enteric bacteria. Failure of therapy of intra-abdominal infections has been noticed more often with chloramphenicoll” and therefore this drug is not recommended. Penicillin is added to metronidazole in the therapy of intracranial and dental infections to “cover” for microaerophilic streptococci, Actinomyces sp. and Arachnia sp. Erythromycin is added to metronidazole in upper respiratory infections to treat S. aureus and aerobic streptococci. Penicillin is added to clindamycin to supplement its coverage against Peptostreptococcus sp. and other grampositive anaerobic organisms. Doxycycline is added to all regimens in the treatment of pelvic infections to provide therapy for chlamydia and mycoplasma. Penicillin is still the drug of choice for bacteremia due to non-BLPB. However, other agents should be used for the therapy of bacteremia due to BLPB. Because the duration of therapy for strict anaerobic infections is generally longer than for infections due to aerobic and facultative anaerobes, which are often chronic, oral therapy is often substituted for parenteral therapy. The agents available for oral therapy are limited and include clindamycin, amoxacillin plus clavulanic acid, chloramphenicol and metronidazole. Clinical judgment, personal experience with the antimicrobial agents, safety, and patient complicance should direct the physician in the choice of the appropriate antimicrobial agents. SingZe Agent or Combined Antimicrobial Therapy?-The principle of using antimicrobial coverage effective against both aerobic and anaerobic pathogens involved in polymicrobial infections has become a cornemtone of practice,16s170 and has been confirmed by numerous studies, especially in intra-abdominal infections.171’172 The success rate in mixed infections varies from study to study, but the difference between various therapeutic regimens is not statistically significant as long as the therapies adequately cover both EnterobacCur-r Probl

Pediatr,

October

1987

606

2 %

z 3 9

=: Jh 0

2

2 3 7 B

P

16.

respiratory

BLPB

with

1. 2. 1. 2. 1. 2. 1. 2.

1. 2. 1. 2.

1. 2. 1. 2. 1. 2. 1. 2.

Infections*

Imipenem

use cefoxitin.

CA = clavukmic

acid;

1. 2. 1. 2. 1. 2. 1. 2.

1. 2. 1. 2.

1. 2. 1. 2. 1. 2. 1. 2.

bacteria,

Clindamycin, Metmnidazoles Clindamycin Chloramphenicol, Clindamycin, Chloramphenicol Penicillin Metronidazole,

MetmnidazoleS: Chloramphenicol Clindamycin, Metmnidazole,* Clindamycin, Chloramphenicol, Clindamycin Chloramphenicol, Amoxicillin + Clindamycin, Chloramphenicol Clindamycinll Amoxicillin +

ORAL

BLPB = Beta-lactamase-producing

Clindamycin,t Cefoxitin,t Metronidazolet Imipenem, Ticarcillin + CA Cefoxitinll Clindamycin,tll Ticarcillin + CA,11 Metmnidazolel/ Clindamycin, cefoxitin Metronidazole+ + Methicillin Clindamycin Chloramphenicol, Imipenem, Metronidazolet Clindamycin, Metmnidazole Cefoxitin, Imipenem Penicillin Clindamycin, Metmnidazole, Cefoxitin

Metronidazole,+

MetmnidazoleQ

Chloramphenicol

Anaerobic

Metronidazole* Chloramphenicol Clindamycin Metmnidazole,+ Clindamycin Chloramphenicol, Clindamycin+ Chloramphenicol,

I?mENTER4L

of Site Specific

NA = not applicable;

and oral areas

drugs;

NA

Cefazolinll Vancomycin CefazolinB Vancomycin NA

Cefoxitin Clindamycint Cefoxitin Doxycycline

Penicillin Vancomycin Penicillin Erythmmycin Cefoxitin Clindamycin NA

‘1 = drug(s) Of choice; 2 = ahemative t = plus aminoglycoside. 9 = plus penicillin. . 5 = plus erythmmycin. II = plus doxycycline. 7 = in locations proximal to the rectal

non

with

Bacteremia

Bone

and joint

1. 2. 1. 2.

2. 1. 2. 1. 2.

1.

Skin

Pelvic

for the Therapy

S”RGICAl PROPHnAxlS

1. 2. 1. 2.

BLPB

tract

Recommended

Abdominal

Pulmonary

Upper

Dental

Intracranial

SITE

Antimicrubials

TABLE

+ CA

Chloramphenicol

Metmnidazole* Metmnidazole

amoxicillin

CAlI

metmnidazole,S CA Metmnidazole

Amoxicillin + CA Chloramphenicol Amoxicillin + CA Metmnidazole*$j

and the B.@agiZis group. A few of these studies used single antimicrobial therapy of a cephalosporin (cefoxitin or moxalactam), and achieved comparable success with combination therapy of either clindamycin or metronidazole plus an aminoglycoside.‘73~20~z03 However, in one of these studies,‘03 single therapy was used only in patients with community-acquired infections and in those who were not immunosuppressed. In those instances, cefoxitin was combined with an aminoglycoside. This was done to avoid failure due to hospital-acquired resistant strains. The use of single-agent therapy was studied more thoroughly in the management of intra-abdominal infection after abdominal trauma. The use of a single antimicrobial agent in trauma is favored because most traumatized patients are young and generally do not have other major medical problems. They usually are admitted to the hospital a short time after the injury, and do not harbor hospitalacquired resistant bacterial strains. Single-agent therapy with cefoxitin was found to be as effective as clindamycin plus an aminoglycoside, 17’,17382oo-zo3 and superior to cefamandole.zO1’ ‘02 Other single drugs or drug combinations that have shown the potential of being effective in the therapy of intra-abdominal infection are imipenem204’ ‘05 and the combination of ticarcillin and clavulanic acid.‘06 Some of the newer cephalosporins promise also to be effective in these infections. However, further studies are needed to ascertain the potential efficacy of these drugs as single therapy of intra-abdominal sepsis. Because of the potential development of bacterial resistance, the use of these newer, highly effective drugs should be/limited to immunocompromised patients or those who developed infection while hospitalized. Single-agent therapy provides the advantage of avoiding the ototoxicity and nephrotoxicity of aminoglycosides, and it is less expensive. But a single agent may not be effective against hospital-acquired resistant bacterial strains, and the use of a single agent lacks antibacterial synergy,2o7 which may be important in immunocompromised hosts. However, for otherwise healthy individuals, when therapy is initiated without a long delay, single agents provide adequate therapy.

teriaceae

Use of Synergistic Antibacterial aerobic Infections.-Combinations

Combinations

in Therapy

of An-

of antibiotics are continually being studied against a variety of infectious agents in attempts to discover more effective therapy for serious infections. In theory, combination therapy might delay emergence of antimicrobial resistance, provide broad spectrum coverage for infections of unknown or mixed etiology, or generate a greater antibacterial effect against specific pathogens than is achievable with either single drug or a combination of two drugs kynergism). The improved effectiveness Cut-r

Probl

Pediatr,

October

1987

607

(as expressed by effective bactericidal activity) of the offending anaerobic organisms is especially important in the treatment of endocarditis and bacteremia or other serious infections. Another situation of concern is the treatment of closed space infections such as brain or lung abscesses that cannot be surgically drained either because of location or the patient’s clinical condition. Most studies on synergistic combinations of antimicrobials were done on Bacteroides fiagilis group (Table 17). Metronidazole has been recognized as one of the most effective antimicrobial agents, consistently inhibitory and bactericidal at achievable in vivo concentrati0ns.l” Because of this finding, that agent has been most frequently studied in combination with other antibiotics, such as clindamycin207-210 and spiramycin,211’217 which proved to be synergistic. The combination of clindamycin and gentamicin has been found to be synergistic by somez*z’213 but not by all investigators. There is general agreement that gentamicin alone is relatively ineffective.18’ Against B. mehinogenicus, the effective combinations were penicillin or clindamycin plus gentamicin, and metronidazole with spiramycin or gentamicin.207~211 Although rare, in vitro and more often in vivo synergy between penicillin, clindamycin or metronidazole and gentamicin against Clostridium sp. and anaerobic cocci does occur. Although such synergy is less likely to be found with gram-positive anaerobic organisms than with Bacteroides sp.,‘16 when present it may offer significant clinical advantages. The in vitro and in vivo synergism between penicillin and gentamicin against B. melaninogenicus is of particular interest. Synergistic interaction between aminoglycosides and penicillins has been noted and studied with certain aerobic or facultative anaerobic organwas found to be effective in isms .‘18 For example, this combination the treatment of enterococcal and staphylococcal diseases. It has been postulated that the penicillins, which inhibit cell wall synthesis, enhance the penetration of aminoglycosides, which are capable of interacting with the ribosomes. There is circumstantial evidence that such a mechanism may prevail in B. melaninogenicus. Bryan et al.218 demonstrated that cell-free amino acid incorporation by B. fragdis ribosomes was inhibited by gentamicin to about the same extent as by E. coli ribosomes. Furthermore, there was no evidence of inactivation of the antibiotic by B. Ji-asilis cell extracts. Whole cells of B. fragilis, however, did not show any time-dependent accumulation of the antibiotic. This failure was attributed to the lack of proper electron transport system for the transport of the aminoglycoside. The mechanism by which penicillin presumably permits the transport of aminoglycosides in Bacteroides sp. has not been investigated. Some of the antimicrobial combinations that showed synergy are used routinely for the therapy of mixed aerobic-anaerobic infections. 99s

Cur-r

Pmbl

Pediatr,

October

1987

TABLE

17.

Summary of Studies Evaluating Synergistic Agents Effective Against Bacteroides fiagilis

Metrunidazole Ampicillin F&unpin Clindamycin Nalidixic acid Erythromycin Spiramycin Gentamicin C&enicillin Cefoxitin Clindamycin Gentamicin Chloramphenicol Cefumxime Penicillin Carbenicillin Mecillinam Carbenicillin Bacteroides

208 209 209,210 209 209,210 211 207 210 210

1X26 (46%) 6/19 (32%)

207,212,213 210

z/3 (66%) 213 (66%) 12/29

melaninogenicus

214 214

(41%)

215

group

3/15 (2O%b)* 2/3 ISSW) 10/15

207 211

(66%)

207

11115 (73%)

207

sp.

Clindamycin Gentamicin

l/l2

Peptostreptococcus

(8%)

216

sp.

Metmnidazole spimmycin Clindamycin Gentamicin ‘Number

of Antimicrobial

gr.

13116 (80%)’ 17/z (77%) 29138 (76%) 14/19 (74%) x/54 (67%) 315 (60%) 7/15 (47%) 4/.28 114%) 3130 (10%)

Metmnidazole Gentamicin Spiramycin Clindamycin Gentamicin Penicillin Gentamicin Clostridium

Combination

7/16 143%) l/7

of isolates where

syoergy

217

(14% 1

was demonstrated/number

216 of bacterial

strains

tested

v% synergyL

These include the combination of clindamycin or metronidazole plus aminoglycosides used for the therapy of intra-abdominal and pelvic infection, and the combination of metronidazole and spiramycin used for the therapy of upper respiratory infections. The synergistic effect against some anaerobic strains produced by these combinations is a valuable additional asset. Curr

Probl

Pediatr,

October

1987

603

The data available so far are restricted to in vitro susceptibility testing and animal studies. Clinical studies in patients are warranted to evaluate the efficacy of synergistic therapy of anaerobic infections. Conclusions.-Polymicrobial infections are generally due to aerobic and anaerobic flora that act synergistically. Appropriate antimicrobial therapy is an important adjunct to surgical management. It should be started as soon as possible, and include either combined or single therapy of antimicrobials effective against the aerobic and anaerobic pathogens. REFERENCES 1. Brook I: Anaerobic infections in childhood. Rev Infect Dis 1984; G(supp1 ljS187. 2. Finegold SM: Anaerobic Bacteria in Human Disease. New York, Academic Press, 1977. 3. Brook I: Anaerobic Infections in Childhood. A Textbook. Boston, GK Hall, 1983. 4. Brook I: Beta-lactamase-producing bacteria recovered after clinical failures with various penicillin therapy. Arch Otolaryngoll984; 110228. 5. Brook I: The role of beta-lactamase-producing bacteria in the persistence of streptococcal tonsillar infection. Ret Infect Dis 1984; 6:601. 6. Swenson RB: Rationale for identification and susceptibility testing of anaerobic bacteria. Rev Infect Dis 1986; 8809. 7. Liu YS, Lim DJ, Lang R, et al: Microorganisms in chronic otitis media with effusion. Ann Otol Rhino1 Laryngoll976; 85245. 8. Mann RJ, Hoffeld TA, Farmer CB: Human bite infection of hand: Twenty years of experience. J Hand Surg 1977; 2:97. 9. Brook I, Finegold SM: Bacteriologv of chronic otitis media. JAM4 1979; 2411487. 10. Brook I: Microbiology of human and animal bite wounds. Pediatr Infect Dis 1987; 629. 11. Pecora DV: A method of securing uncontaminated tracheal secretions for bacterial examination. J Thorac Surg 1959; 37:653. 12. Brook 1: Percutaneous transtracheal aspiration in the diagnosis and treatment of aspiration pneumonia in children. J Pediatr 1980; 9O:lOOO. 13. Bartlett JG, Rosenblatt JE, Finegold SM: Percutaneous transtracheal aspiration in the diagnosis of anaerobic pulmonary infection Ann Intern Med

1973; 22:535. 14. Spenser CD, Beaty Med 1972; 286:304. 15.

16. 17. 18. 19. 610

HN:

Complications

of transtracheal

aspiration.

N Engl J

Dowell VR, Jr: Anaerobic infections, in Bodily HL, Updyke EL, Mason JO teds): Diagnostic Procedures for Bacterial, Mycotic and Parasitic Infections, ed 5. New York American Public Health Association, 1970, pp 494. Holdeman LV, Moore WEC (ed): Anaerobe laboratory Manual, ed 4. Blacksburg, Va, Virginia Polytechnic Institute and State University, 1977. Sutter VL, Citron DM, Edelstein MAC, et al: Anaerobic Bacteriology Manual, ed 4. Belmont, Calif, Star Publishing Co, 1985. Loesche WJ: Oxygen sensitivity by various anaerobic bacteria. Appl Environ Microbial 1973; 18:911. Hankle ME, Katz YJ: An electrolytic method for controlling oxidation-

Curr Probl Pediatr, October

1987

20. 21. 22. 23. 24. 25.

26.

reducing

potential

Biochem

Biophys

and its application in the study of anaembiosis. Arch 1943; 2:183. Tally FP, Stewart PR, Sutter VL, et al: Oxygen tolerance of fresh clinical anaerobic bacteria. J Clin Microbial 1975; 1:161. Syed SA, Loesche WJ: Survival of human dental plaque flora in various transport media. Appl Environ Microbial 1972; 24:638. Mena E, Thompson ,FS, Armfield AY, et al: Evaluation of Port-A-Cul transport system for protection of anaerobic bacteria. J Clin Microbial 1978; 828. McConville JH, Timmons RF, Hansen SL: Comparison of three transport systems for recovery of aembes and anaembes from wounds. Am J Clin Path01 1979; 72:968. Tally FP, Cuchural GJ, Jacobus NV, et al: Susceptibility of Bacteroides fragilis gmup in the United States in 1981. Antimicrob Agents Chemother 1983; 23336. Marrie TJ, Haldane EV, Swantee CA, et al: Susceptibility of anaerobic bacteria to nine antimicrobial agents and demonstration of decreased susceptibility of Clostridium perji-ingens to penicillin. Antimicrob Agents Chemother 1981; 1951. Bourgault A-M, Rosenblatt JE: Characterization of anaerobic gram-negative bacilli by using rapid slide tests for beta-lactamase-production. J Clin Microbiol

1979;

9:654.

27. Rosenblatt 28. 29. 30. 31.

JE: Antimicrobial susceptibility testing of anaerobic bacteria. Rev h$xt Dis 1984; 6(suppl):S242. Bmok I, Finegold SM: Aerobic and anaerobic bacteriology of cutaneous abscesses in children. Pediatrics 1981; 67:891. Brook I, Randolph JG: Aerobic and anaerobic bacterial flora of burns in children. J Trauma 1981; 21:313. Bmok I: Bacterial studies of peritoneal cavity and postoperative surgical wound drainage following perforated appendix in children. Ann Surg 1980; 192:208. Bmok I: Anaerobic and aerobic bacteriology of decubitus ulcers in children.

Am

Surg

32. Brook Am

1980;

46:624.

I, Contmni

G, Rodriguez WJ, et al: Anaerobic bactermia in children. 134:1052. I: Bacteriology of intracranial abscess in children. J Neurosurg 1981;

J Dis

Child

1980;

33. Bmok 54:484. 34. Brook I, Friedman EM, Rodriguez WJ, et al: Complications of sinusitis in children. Pediatrics 1980; 66:568. 35. Brook I: Bacteriologic features of chronic sinusitis in children. JAA4A 1981; 246:967. 36. Brook I: Aerobic and anaerobic bacteriology of chronic mastoiditis in children. Am J Dis Child 1981; 135:478. 37. Brook I: Otitis media in children: A prospective study of aerobic and anaerobic bacteriology. Laryngoscope 1979; 89:992. 38. Brook I: Chronic otitis rhedia in children. Microbiological studies. Am J Dis Child 1980; 134:564. 39. Brook I, Finegold SM; Bacteriology of aspiration pneumonia in children. Pediatrics

1980;

65:1115.

40. Brook I, Finegold J Pediatr

1979;

SM: Bacteriology

and therapy

41. Brook

I, Gluck RS: Clostridium paraputr$cum J 1980; 73:%X4. 42. Brook I, Schwartz RH, Contmni G: Clostridium South

Curr

Probl

of lung abscess in children.

94:10.

sepsis in sickle cell anemia.

Med

Pediatr,

October

1987

ramosum

and beta-hemo611

43. 44.

45. 46. 47.

48.

49. 50. 51. 52. 53. 54.

55. 56. 57.

58. 59. 60. 61. 62. 63. 64. 65.

612

lytic streptococci isolated from a child presenting with acute otitis media. Clin Fed&r (Phila) 1979; 18699. Brook I: Aerobic and anaerobic bacteriology of peritonsillar abscess in children. Acta Paediatr Stand 1981; 70:831. Brook I, Martin WJ, Finegold SM: Effect of silver nitrate application on the conjunctival flora of the newborn and the occurrence of clostridial conjunctivitis. J Pediatr Ophthalmol Strabismus 1978; 15:179. Brook I: Bacteriology of neonatal omphalitis. J Infect 1982; 5:127. Howard FM, Flynn DM, Bradley JM, et al: Outbreak of necrotizing enterocolitis caused by Clostridium butyricum. Lancer 1977; 2(8048):1099. Sturm R, Staneck JL, Stauffer LR, et al: Neonatal necrotizing enterocolitis associated with penicillin-resistant, toxigenic Clostridium butyricum. Fediatrics 1980; 66:928. Brook I: Isolation of toxin producing Clostridium d@cile from two children with oxacillinand dicloxacillin-associated diarrhea. Pediatrics 1980; 65:1154. Viscidi RP, Bartlett JG: Antibiotic-associated pseudomembranous colitis in children. Pediatrics 1981; 67:381. Brook I, Avery G, Glasgow A: Clostridium dtjkile in pediatric infections. J Infect 1981; 4:253. Fischer GW, Sunakorn P, Duangman C: Otogenous tetanus: A sequelae of chronic ear infections. Am J Dis Child 1977; 131445. Tally FP, Armfield AY, Dowel1 VR Jr, et al: Susceptibility of Clostridium ramosum to antimicrobial agents. Antimicrob Agents Chemother 1974; 45:589. Drake DP, Holt RJ: Childhood actinomycosis: Report of three recent cases. Arch Dis Child 1976; 51:979. Brook I, Barrett CT, Brinkman CR, et al: Aerobic and anaerobic bacterial flora of maternal cervix and newborn gastric fluid and conjunctiva: A prospective study. Pediatrics 1979; 63:4.51. Brook I, Martin WJ, Finegold SM: Neonatal pneumonia caused by members of the Bacteroidesfragilis group. Clin Pediatr (Phila) 1980; 19:541. Brook I: Osteomyelitis and bacteremia caused by Bacteroides fragilis: A complication of fetal monitoring. Clin Pediatr (Phila) 1980; 19:639. Aldridge KE, Sanders CV, Janney A, et al: Comparison of activities of penicillin G and new beta-lactam antibiotics against clinical isolates of Bacteroides species. Antimicrob Agents Chemother 1984; 26:410. Brook I: Bacteriologic study of paronychia in children. Am J Surg 1981; 141:703. Brook I: Urinary tract infections caused by anaerobic bacteria in children. Urology 1980; 16:596. Brook I, Grimm S, Kielich RB: Bacteriolo@ of acute periapical abscess in children. J Endodon 1981; 7:378. Brook I, Calhoun L, Yocum P: Beta-lactamase-producing isolates of Bacteroides species from children. Antimicrob Agents Chemother 1980; 18264. Brook I, Yocum P, Shah K: Surface vs. core-tonsillar aerobic and anaerobic flora in recurrent tonsillitis. JAM4 1980; 244:1696. Brook I: Prevalence of beta-lactamase-producing bacteria in chronic suppurative otitis media. Am J Dis Child 1985; 139:280. Eschenbach OA: Acute pelvic inflammatory disease. Ural Clin North Am 1984; 11:65. Tuner K, Lindqvist L, Nord CE: Purification and properties of a novel j3lactamase from Fusobacterium nucleatum. Antimicrob Agents Chemother 1985; 27:943. Curr

Probl

Pediatr,

October

1987

66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84.

Tedesco FJ, Barton RW, Alpers DH: Clindamycin-associated colitis: A pmspective study. Ann Intern Med 1974; 81:429. Pedensar PV, Hansen FH, Halveg AP, et al: Necmtizing entemcolitis of the newborn-is it gas-gangrene of the bowel? Lancet 1976; 2:715. Kliegman RM, Faanmff AA, Izant RJ, et al: Clostridia as pathogens in neonatal necrotizing enterocolitis. J Pediatr 1979; 95287. Long SS, Swenson ,RM: Development of anaerobic fecal flora in healthy newborn infants. J Pediatr 1977; 91:298. Chang TW, Areson P: Neonatal necmtizing entemcolitis: Absence of enteric bacterial toxin. N Engl J Med 1978; 299:424. Brook I: Aerobic and anaerobic bacterial isolates of acute conjunctivitis in children: A prospective study. Arch Ophthalmol 1980; 98:833. Matsuura H: Anaembes in the bacterial flora of the conjunctival sac. Jpn J Ophthalmol 1971; 15:116. Brook I, Pettit TH, Martin WJ, et al: Aerobic and anaerobic bacteriology of acute conjunctivitis. Ann Ophthalmology 1978; 11:X3. Brook I: Recovery of anaerobic bacteria in conjunctivitis associated with wearing contact lens. Submitted for Publication. Rosebury T: Microorganisms Indigenous to Man. New York, MacGraw-Hill, 1962. Brook I: Role of anaerobic bacteria in otitis media: Microbiology, pathogenesis and implications on therapy. Am J Otolaryngol 1987; 8:109. Brook I, Schwartz R: Anaerobic bacteria in acute otitis media. Acta Otolaryngol 1981; 91:lll. Brook I, Yocum P, Shah K, et al: Aerobic and anaerobic bacteriological features of serous otitis media in children. Am J Otolaryngol 1983; 4:389. Karma P, Jokipii L, Ojala K, et al: Bacteriology of the chronically discharging middle ear. Acta Otolaryngol 1978; 86:llO. Sugita R, Kawamura S, Ichikawa C, et al: Studies of anaerobic bacteria in chronic otitis media. Laryngoscope 1981; 9:816. Aygagari A, Pancholi VK, Pandhi SC, et al: Anaerobic bacteria in chronic suppurative otitis media. Indian J Med Res 1981; 73:860. Sweeney G, Picozzi GL, Bmwning GG: A quantitative study of aerobic and anaerobic bacteria in chronic suppurative otitis media. J Infect 1982; 5:47. Constable L, Butler I: Microbial flora in chronic otitis media. J kfect 1982; 5:57. Papastavms T, Giamarelou H, Varlejides S: Role of aerobic and anaerobic microorganisms in chronic suppurative otitis media. Laryngoscope 1986;

96:438. 85. 86.

87.

Brook I: Bacteriology and treatment of chronic otitis media. 1979; 89:1129. Kenna MA, Bluestone CO, Reilly .I, et al: Medical management suppurative otitis media without cholesteatoma in children. 1986; 96:146. Brook I: Aerobic and anaerobic bacteriology of cholesteatoma.

Laryngoscope of chronic

Laryngoscope Laryngo-

scope 1981; 91250. 88.

89.

90. Burr

Iino Y, Hoshimi E, Tomioko S, ganisms in the contents of the 1983; 92:91. Brook I, Yocum P: Comparison cal and non-Group A stmptococcal press. Bmok I, Yocum P, Friedman Probl

Pediatr,

October

1987

et al: Organic cholesteatoma of the

EM:

acids sac.

microbiology tonsillitis.

Aerobic

and

and

anaerobic

micmor-

Ann Otol Rhino1 Laryngol of Gmup

A streptococ-

Ann Otol Rhino1 kryngol, anaembic

flora

in

recovered 613

91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 614

from tonsils of children with recurrent tonsillitis. Ann Otol Rhino1 Laryngol 1981; 90261. Reilly S, Imms P, Beeden AG, et al: Possible role of the anaerube in tonsillitis. J Clin Pathol 1981; 34:542. Tuner K, Nord CE: Beta-lactamase-producing microorganisms in recurrent tonsillitis. Stand J Infec Dis 1983; 39(suppl):83. Chagollan JJ, Ramirez MJ, Gil JS: Flora indigena de las amigdalas. Investigacion Medica lnternacional 1984; 11:36. Brook I: Microbiology of retropharyngeal abscesses in children. Am J Dis Child 1987; 141202. Stammers AF: Vincent’s infection: Observation and conclusions regarding the aetiology and treatment of 1017 civilian cases. Br Dent J 1944; 76:147. Brook I, Gober AE: Bacteroides melaninogenicus: Its recovery from tonsils of children with acute tonsillitis. Arch Otolaryngol 1984; 109:818. Davidson S, Kaplinsky C, Frand M, et al: Treatment of infectious mononucleosis with metronidazole in the pediatric age group. Stand J Infect Dis 1982; 14:103. Helstrom SA, Mandl PA, Ripa T: Treatment of anginose mononucleosis with metronidazole. Stand J kQixt Dis 1978; 10:7. McDonald CJ, Tierney WM, Hui SL: A controlled trial of erythrumycin in adults with non streptococcal pharyngitis. J Infect Dis 1985; 152:1093. Merenstein JH, Rogers KD: Streptococcal pharyngitis. Early treatment and management by nurse practitioners. JAMA 1974; 227:1278. Putto A: Febrile exudative tonsillitis: viral or streptococcal. Pediatrics 1987; 806. Tally FP, Sutter VL, Finegold SM: Treatment of anaerobic infections with metronidazole. Antimicrob Agents Chemother 1975; 7:672. Brook I, Fink R: Transtracheal aspiration in pulmonary infection in children with cystic fibrosis. Eur J Respir Dis 1983; 64:51. Thomassen MJ, Klinger JD, Badger SJ, et al: Cultures of thoracotomy specimens confirm usefulness of sputum cultures in cystic fibrosis. J Pediatr 1986; 104:352. Gorbach SL, Bartlett JG: Anaerobic infections. N Engl J Med 1974; 290:1177. Mergenhagen SE, Thonard JC, Scherp HW: Studies on synergistic infection. I: Experimental infection with anaerobic streptococci. J Infect Dis 1958; 103:33. Lev M, Krudell KC, Milford AF: Succinate as a growth factor for Bacteroides melaninogenicus. J Bacterial 1971; 108:175. Meleney FL: Bacterial synergy in disease processes. Ann Surg 1931; 22:961. Altemeier WA: The pathogenicity of the bacteria of appendicitis. Surgery 1942; 11:374. Meleney FL, Olpp S, Harvey HD, et al: Peritonitis. II: Synergism of bacteria commonly in peritoneal exudates. Arch Surg 1932; 25:709. Hite KE, Locke M, Heseltine HC: Synergism in experimental infections with nonsporulating anaerobic bacteria. J Infect Dis 1949; 84:l. Brook I, Hunter V, Walker Ri: Synergistic effects of anaerobic cocci, Bacteroides, Clostridia, Fusobacteria, and aerobic bacteria on mouse mortality and induction of subcutaneous abscess. J Infect Dis 1984; 149924. Brook I, Walker RI: Significance of encapsulated Bacteroides melaninogenicus and Bacteroidesfiagilis groups in mixed infections. Infect Immun 1984; 44:12. Brook I: Enhancement of growth of aerobic and facultative bacteria in Curr Probl Pediatr, October

1987

mixed

infections with Bacteroides fragilis and melaninogenicus groups. In1985; 50:929. Brook I: Enhancement of growth of aerobic, anaerobic and facultative bacteria in mixed infections with anaerobic and facultative gram positive cocci. J Surg Res, in press. Brook I, Walker RI: The relationship between Fusobacterium species and other flora in mixed infection. J Med Microbial 1986; 21:93. Brook I, Walker RI: Pathogenic@ of some Clostridium species. J Infect 1986; 13245. Ingham HR, Sisson PR, Tharagonnet D, et al: Inhibition of phagocytosis in vitro by obligate anaerobes. Lancet 1977; 2:1251. Onderdonk AB, Cisneros DL, Bartlett JG: The capsular polysaccharide of Bacteroides fiagilis as a virulence factor: Comparison of the pathogenic potential of encapsulated strain. J Infect Dis 1977; 136:82. Simon GL, Klempner MS, Kasper DL, et al: Alterations in opsonophagocytic killing by neutrophils of Bacteroides fkgilis associated with animals and laboratory passage: Effect of capsular polysaccharide. J Infect Dis 1982; 145172. Okunda K, Takazoe I: Antiphagocytic effects of the capsular structure of a pathogenic strain of Bacteroides melaninogenicus. Bull Tokyo Med Den

fect

115. 116. 117. 118. 119. 120.

121.

Univ

Immun

1973;

14:99.

122. Tofte RW, Peterson PK, Schmeling D, et al: Opsonization of four Bacteroides species: Role of the classical complement pathway and immunoglobulin. Infect Immun 1980; 27:784. 123. Jones GR, Gemmel CG: Impairment by Bacteroides species of opsonization and phagocytosis of enterobacteria. J Med Microbial 1982; 15:351. 124. Patrick S, Reid JH, Larkin MJ: The growth and survival of capsulate and non-capsulate Bacteroides fi-agilis in vivo and in vitro. J Med Microbial 1984;17:237. 125. Rotstein OD, Pruett TL, Fiegel CD, et al: Succinic acid, a metabolic by-product of Bacteroides species, inhibits polymorphonuclear leukocytes function. Infect Immun 1985; 48:402. 126. Brook I, Gillmore JD, Coolbaugh JC, et al: Pathogenic@ of encapsulated Bacteroides melaninogenicus group, Bacteroides oralis, and Bacteroides ruminicola in abscesses in mice. J Infect 1983 7218. 127. Brook I, Walker RI: Pathogenic@ of anaerobic gram positive cocci. Infect Immun 1984;45:320. 128. Brook I, Walker RI: Infectivity of organisms recovered from polymicmbial abscesses. Infect Immun 1983; 41:986. 129. Brook I: Bacteremia and seeding of encapsulated Bacteroides sp. and anaerobic cocci. J Med Microbial 1987; 23:61. 136. Wood WB, Smith MR: Inhibition of surface phagocytosis by the capsular slime layer of Pneumococcus type III. J E,xp Med 1949; 90:86. 131. Chandler CA, Fothergill LD, Dingle JH: Studies of Haemophilus influenzae. II: A comparative study of virulence of smooth, rough and respiratory strains of Haemophilus infhtenzae as determined by infection of mice with mucin suspension of the organism. J Exp Med 1937; 66:789. 132. Brook I: Recovery of encapsulated anaerobic bacteria from orofacial abscesses. J Med Microbial 1986; 22:171. 133. Scheifele DW, Fussell SJ: Ampicillin-resistant Haemophilus infruenzae colonizing ambulatory children: Epidemiology and implications for otitis media therapy. Am J Dis Child 1981; 135:406. Curr

Probl

Pediatr,

October

1987

615

134. Kovatch

AL, Wald ER, Michaels RH: Beta-lactamase-producing Branhamella causing otitis media in children. J Pediatr 1983; 102260. 135. Murray PR, Rosenblatt JE: Penicillin resistance and penicillinase pmduction in clinical isolates of Bacteroides melaninogenicus. Antimicrob Agents catarrhalis

Chemother

1977;

11:605.

136. Brook I: Microbiology of abscesses of head and neck in children. Ann Otol Rhino1 Laryngol, in press. 137. Brook I, Martin WJ: Aerobic and anaerobic bacteriology of perirectal abscess in children. Pediatrics 1980; 66282. 138. Brook I, Anderson K, Rodriguez WJ, et al: Aerobic and anaerobic bacteriology of pilonidal cyst -abscess in children. Am J Dis Child 1980; 134629. 139. Brook I: Aerobic and anaerobic bacteriology of cervical adenitis in children. Clin

Pediatr

1980;

19:693.

140. Bmok I: Aerobic and anaerobic bacteriology of adenoids in children: Comparison between patients with chronic adenotonsillitis and adenoid hypertrophy. laryngoscope 1981; 91:377. 141. Brook I, Yocum P: Bacteriology of chronic tonsillitis in young adults. Arch Otolaryngol 1984; 110:803. 142. Bmok I: Bacterial colonization, trachitis and pneumonia, following tracheostomy and long-term intubation in pediatric patients. Chest 1979; 70:420. 143. Brook I: Presence of beta-lactamase-producing bacteria and beta-lactamase activity in abscesses. Am J Path01 1986; 8697. 144. Brook I, Altman RP: The significance of anaerobic bacteria in biliary tract infections following hepatic porto-entemstomy for biliary atresia. Surgery 1984; 95281. 145. Bmok I: Anaerobic osteomyelitis in children. Pediatr k$ixt Dis 1986; 5550. 146. Hackman AS, Wilkins TD: In vivo protection of Fusobacterium necrophorum iiom penicillin by Bacteroides fiagilis. Antimicrob Agents Chemother 1975; 7:698. 147. Brook I, Pazzaglia G, Coolbaugh JC, et al: In vitm protection of Group A beta-hemolytic streptococci by beta-lactamase-producing Bacteroides species. J Antimicrob Agents Chemother 1983; 12599. 148. Simon HM, Sukai W: Staphylococcal pharyngeal infection: Effect of oxacillin. Pediatrics 1963; 31:463. 149. Scheifele DW, Fussell SJ: Frequency of ampicillin resistant Haemophilus parainjhrenzae in children. J Infect Dis 1981; 143:495. 150. Brook I, Yocum P: In vitro protection of Group A beta-hemolytic streptococci from penicillin and cephalothin by Bacteroides jiagilis. Chemotherapy

1983;

29:lS.

151. de Louvois J, Hurley R: Inactivation of penicillin by purulent exudates. Br Med J 1977; 2:998. 1.52. Matsuda G, Tomioka S: Possible beta-lactamase activities detectable in infective clinical specimens. J Antibiot (Tokyo1 1977; 30:1093. 153. O’Keefe JP, Tally FP, Barza M, et al: Inactivation of penicillin G during experimental infection with Bacteroides fragilis. J Infect Dis 1978; 137:437. 154. Bryant RE, Rashad AL, Mazza JA, et al: Beta-lactamase activity in human pus. J Infect Dis 1980; 142:594. 155. Boughton WH: Rapid detection in spinal fluid of beta-lactamase produced by ampicillin-resistant Haemophihzs injhrenzae. J Clin Microbial 1982; 25:1167. 156. Yolken RH, Hughes WT: Rapid diagnosis of infections caused by beta-lactamase-producing bacteria by means of an enzyme radioisotopic assay. J Pediatr 616

1989;

97:715. Curr

Probl

Pediatr,

October

1987

1.57. Heimdahl A, Von Konow L, Nord CE: Isolation of beta-lactamase-producing Bacteroides strains associated with clinical failures with penicillin treatment of human orofacial infections. Arch Oral Bioll980; 25288. 1.58. Gastanaduy AS, Kaplan EL, Huwe BB, et al: Failure of penicillin to eradicate Group A streptococci during an outbreak of pharyngitis. Lancet 1980; 2498. 159. Knudsin RB, Miller JM: Significance of the Staphylococcus aureus carrier state in the treatment of disease due to Group A streptococci. New Engl J Med 1964; 271:1395. 160. Quie PG, Pierce AX, Wannamaker LW: Influence of penicillinase producing staphylococci on the eradication of Group A streptococci from the upper respiratory tract by penicillin treatment. Pediatrics 1966; 37:467. 161. Bmok I, Himkawa R: Treatment of patients with a history of recurrent tonsillitis due to Gmup A beta-hemolytic streptococci. Clin Pediatr 1985; 24:331. 162. Brook I, Yocum P: Quantitative measurement of beta-lactamase levels in tonsils of children with recurrent tonsillitis. Acta Otolarpgol Stand 1984; 98:446.

163. Bmok I, Gober AE: Emergence of beta-lactamase-producing aerobic and anaerobic bacteria in the ompharynx of children following penicillin chemotherapy. C/in Pediatr 1984; 23:338. 164. Tuner K, Nord CE: Emergence of beta-lactamase-producing micmorganisms in the tonsils during penicillin treatment. Eur J Clin Microbial 1986; .5:399.

165. Brook I: Emergence and persistence of beta-lactamase-producing bacteria in the ompharynx following penicillin treatment. Submitted for publication. 166. Bmok I: Role of beta-producing bacteria in penicillin failure to eradicate Group A streptococci. Pediatr Infect Dis 1985; 4:491. 167. Ross K, Grahn E, Holm SE: Evaluation of beta-lactamase-activity and microbial interference in treatment failures of acute streptococcal tonsillitis. Stand J Infect Dis 1986; 18:313. 168. Weinstein WM, Onderdonk AB, Bartlett JG, et al: Experimental intraabdominal abscesses in rats. I: Development of an experimental model. Infect

Immun

1984;

10:1250.

169. Thadepalli H, Gorbach SL, Bmido PW, et al: Abdominal trauma, anaembes and antibiotics. Surg Gynecol Obstet 1973; 137~270. 170. Finegold SM, George WL, Mulligan ME: Anaerobic infections. DM 1985; 31:11-12. 171. Bartlett JG: Recent developments in the management of anaerobic infection. Rev Infect Dis 1983; 5:235. 172. Solon&in JS, Meakins JL, Al10 MD, et al: Antibiotic trials in intra-abdominal infections: A critical evaluation of study design and outcome reporting. Ann Surg 1984;200:29. 173. Nichols RL, Smith JW, Klein DB, et al: Risk of infection after penetrating abdominal trauma. N Engl J Med 1984; 311:1065. 174. Geddes AM, Stille W: Imipenem. The first thienamycin antibiotic. Rev Infect Dis 1985; 7:S353. 175. Breese BB, Disney F& Talpey WB: Beta-hemolytic streptococcal illness: Comparison of lincomycin, ampicillin, and potassium penicillin G in treatment. Am J Dis Child 1966; 11221. 176. Breese BB, Disney FA, Talpey WB, Green J: B-hemolytic streptococcal infection: Comparison of penicillin and lincomycin in the treatment of recurrent infections or the carrier state. Am J Dis Child 1969; 117:147. Curr

Probl

Pediatr,

October

1987

617

177.

178.

179.

180.

181.

Randolph MF, DeHaan RM: A comparison of lincomycin and penicillin in the treatment of Group A streptococcal infections: Speculation on the “L” forms as a mechanism of recurrence. Del Med J 1969; 41:51. Howie VM, Plousard JH: Treatment of Group A streptococcal pharyngitis in children: Comparison of lincomycin and penicillin G given orally and benzathine penicillin G given intramuscularly. Am J Dis Child 1971; 121:477. Randolph MF, Redys JJ, Hibbard EW: Streptococcal pharyngitis. III: Streptococcal recurrence rates following therapy with penicillin or with clindamycin (7-chlorolincomycin). Del Med J 1970; 4287. Stillerman M, Isenberg HD, Facklam RR: Streptococcal pharyngitis therapy: Comparison of clindamycin palmitate and potassium phenoxymethyl penicillin. Antimicrob Agents Chemother 1973; 4:516. Masse11 BF: Prophylaxis of streptococcal infection and rheumatic fever: A comparison of orally administered clindamycin and penicillin. JAMA 1979; 241:1589.

182. 183.

184. 185.

186. 187. 188. 189. 190.

191. 192. 193. 194. 195. 196. 197. 198. 618

Brook I, Leyva F: The treatment of the carrier state of Group A beta-hemolytic streptococci with clindamycin. Chemotherapy 1981; 27:360. Chaudhary S, Bilinsky SA, Hennessy JL, et al: Penicillin V and rifampin for the treatment of Group A streptococcal pharyngitis: A randomized trial of 10 days penicillin vs. 10 days penicillin with rifampin during the final 4 days of therapy. J Pediatr 1985; 106:481. Tanz RR, Shulman ST, Barthel MJ, et al: Penicillin plus rifampin eradicate pharyngeal carrier of Group A streptococci. J Pediatr 1985; 106:876. Levison ME, Mangura CT, Lorber B, et al: Clindamycin compared with penicillin for the treatment of anaerobic lung abscess. Ann Inter-m Med 1983; 98:466. Shupak A, Halpern P, Ziser A, et al: Hyperbaric oxygen therapy for gas gangrene casualties in Lebanon War, 1982. Isr J Med Sci 1984; 20:323. Barsoum AH, Lewis HC, Cannillo EL: Non-operative treatment of multiple brain abscess. Surg Neural 1981; 16283. Verklin RM, Mandell GL: Alteration of antibiotics by anaerobiosis. J lab C/in Med 1977; 89:65. Sutter VI,, Finegold SM: Susceptibility of anaerobic bacteria to 23 antimicrobial agents. J Antimicrob Agents Chemother 1976; 10:736. Reading C, Cole M: Clavulanic acid: A beta-lactamase-inhibiting betalactam from Streptomyces clavuligerus. Antimicrob Agents Chemother 1977; 11:852. Wust J, Wilkins TD: Effect of clavulanic acid on anaerobic bacteria resistant to beta-lactam antibiotics. Antimicrob Agents Chemother 1978; 13:130. Heseltine PN, Busch DF, Meyer RD, et al: Cefoxitin: Clinical evaluation in thirty-eight patients. Antimicrob Agents Chemother 1977; 11:427. Chow AW, Montgomerie JZ, Guze LB: Parenteral clindamycin therapy for severe anaerobic infection. Arch Intern Med 1974; 134178. Randolph MF, Morris KE: Clindamycin-associated colitis in children. A prospective study and a negative report. Clin Pediatr (Philal 1977; 16:722. Berlatzky Y, Rubin SZ, Michel J, et al: Use of clindamycin and gentamicin in pediatric colonic surgery. J Pediatr Surg 1976; 11:943. Brook I: Clindamycin in treatment of aspiration pneumonia in children. Antimicrob Agents Chemother 1979; 15:342. Brook I: Treatment of anaerobic infections in children with metronidazole. Dev Pharmacol Ther 1983; 6:187. Ingham HR, Selkon JB, Roxby CM: Bacteriological study of otogenic cerebral Cur-r

Probl

Pediatr,

October

1987

abscesses:

Chemotherapeutic

role

of metronidazole.

Br Med

J

1977;

2(6093):991.

199. Thadepalli A, Gorbach SL, Bartlett JG: Apparent failure of chloramphenicol in anaerobic infections. Obstet Gynecol Surg 1978; 35:334. ZOO. Hofstetter SR, Pachter HL, Bailey AA, et al: A prospective comparison of two regimens of prophylactic antibiotics in abdominal trauma: Cefoxitin versus triple drug. J Trauma 1984; 24:307. 201. Gentry LO, Feliciano DV, Scott AL, et al: Perioperative antibiotic therapy for penetrating injuries of the abdomen. Ann Surg 1984; 200561. 202. Jones RC, Thal ER, Johnson NA, et al: Evaluation of antibiotic therapy following penetrating abdominal trauma. Ann Surg 1985; 201:576. 203. Tally FP, McGowan V, Kellum JM, et al: A randomized comparison of cefoxitin with or without amikacin and clindamycin plus amikacin in surgical sepsis. Ann Surg 1981; 193:318. 204. Schreiner A, Olsen T, Madsen ST, et al: Imipenemcilastatin versus gentamicin/clindamycin for treatment of serious bacterial infections. Lancet 1984; 1:868. 205. Solomkin JS, Fant WK, Rivera JO: Randomized trial of imipenemcilastatin versus gentamicin and clindamycin in mixed flora infection. Am J Med 1985; 78(S6A):85. 206. Leigh DA, Phillips F, Wise R: Timentin-ticarcillin plus clavulanic acid. A laboratory and clinical prospective study. J Antimicrob Chemother 1986; 17(Supp 6):l. 207. Bmok I, Coolbaugh JC, Walker RI, et al: Synergism between penicillin, clindamycin or metmnidazole and gentamicin against species of Bacteroides melaninogenicus and j?agilis groups. Antimicrob Agents Chemother 1984; 25:71. 208. Bergan T, Fotland MH: In vitro interactions between metronidazole or tinidazole and co-trimoxazole or the effect against anaerobic bacteria. &and J Gastroenterol 1984; 19(suppl):95. 209. Ralph ED, Amatnieks YE: Potentially synergistic antimicrobial combinations with metmnidazole against Bacteroides fragilis. Antimicrob Agents Chemother 1980; 17:379. 210. Busch DF, Sutter VL, Finegold SM: Activity of combinations of antimicrobial agents against Bacteroides ji-agihs. J Infect Dis 1976; 133:321. 211. Bmok I: Activity of metronidazole and spiramycin in Bacteroides sp. and Staphylococcus aureus abscesses. Proceedings of the Interscience Pmceedings of the 26th Conference of Antimicrobial Agents and Chemotherapy (abstract no. 551). J Antimicrob Chemother 1986; in press. 212. Fass RJ, Rotilie CA, Prior RB: Interaction of clindamycin and gentamicin in vitro. Antimicrob Agents Chemother 1974; 6:582. 213. Okubadejo OA, Allen J: Combined activity of clindamycin and gentamicin on Bacteroides fragilis and other bacteria. J Antimicrob Chemother 1975; 1:403. 214. Thadepalli H, White DW, Bach VI: Antimicrobial activity and synergism of cefumxime on anaerobic bacteria. Chemotherapy 1981; 271252. 215. Trestman I, Kaye D, Levinson ME: Activity of semi synthetic penicillins and synergism with mecillinam against Bacteroides species. Antimicrob Agents Chemother

1979;

16283.

216. Bmok I, Walker RI: Interaction between penicillin, clindamycin or metmnidazole and gentamicin against species of clostridia and anaerobic and facultative anaerobic Gram-positive cocci. J Antimicrob Chemother 1985; 15:31. Curr

Probl

Pediatr,

October 1987

619

217.

218.

Videau D, Blanchard JC, Sebald M: Antibiotic susceptibility and value of the combination of spiramycin and metronidazole. Ann Microbial 1973; 1248:505. Bryan LE, Kowand SK, Van Den Elzen HM: Mechanism of aminoglycoside antibiotic resistance in anaerobic bacteria: Clostridium per-$-ingens and Bacteroides frasilis. Antimicrob Agents Chemother 1979; 15:7-13.

Curr

Probl

Pediatr,

October

1987