Antimicrobial agents are the most important therapeutic tools for most primary care pediatrie providers. The ready availability of antimicrobials has dramatically changed outpatient pediatrie practice because antimicrobials are relatively inexpensive, largely nontoxic, and, most important, effective in rapidly improving clinical symptoms in many ambulatory infections. As a result, pediatricians and other primary care providers have used antimicrobials liberally, knowing that the benefits of therapy greatly exceed the risk for any individual child.
The magnitude of physician prescribing of antimicrobials is well documented in the current literature, as is the number of prescriptions written for questionable indications. It is estimated that one of every six patients visiting a primary care provider receives an antimicrobial: from 1990 through 1992, 17 million prescriptions were written for upper respiratory tract infections, whereas 16 million and 13 million prescriptions were written for bronchitis and pharyngitis, respectively.1 In one study of family physicians, 60% of patients diagnosed with an upper respiratory tract infection were prescribed an antimicrobial.2 When 346 pediatricians and family physicians were presented the scenario of a 1year-old child with a 1-day history of a mucopumlent nasal discharge, 71% of family physicians and 53% of pediatricians would prescribe an antibiotic. Only 15% of family physicians and 23% of pediatricians would wait 7 to 10 days before beginning therapy. The reasons cited were pressure from family and fear of development of acute otitis media (AOM) in the untreated child.3
Although many factors contribute to the development of antimicrobial resistance, the most consistent risk factor is the intensity of prior antimicrobial exposure.4 Pediatrie providers now encounter significant and increasing rates of antimicrobial resistance among common pediatrie pathogens, including Haemophilus influenzae, Streptococcus pneumoniae, Staphylococcus aureus, and Eschenchia coli.
As a result, the selection of the most appropriate antimicrobial is often difficult. The practitioner must discern the bacteria most likely to cause a given infection and the typical pattern of its antimicrobial susceptibility. Local data on antimicrobial susceptibility are often unavailable to office-based clinicians and when data are available, they may reflect trends for hospitalized patients or patients who are referred to specialty care and may not adequately represent antibiotic sensitivities in ambulatory practice. The rapidly increasing cost of pharmaceuticals, including some antimicrobials, has led managed care organizations and other institutions to restrict prescribing choices by closing formularies or requiring prior approval for certain drugs. Finally, parents often have strong expectations that their child should receive an antimicrobial, and may indicate which one they would prefer.5 Parents are most interested in convenient dosing schedules, an absence of side effects, and minimizing the out-of-pocket costs of their prescriptions. Not surprisingly, the primary care physician may find it difficult to balance these competing interests.
This article briefly reviews common pediatrie pathogens, their usual patterns of antimicrobial susceptibility, and the in vitro spectrum and extent of activity of some currently available antimicrobials against these organisms. Options for therapy for common pediatrie infections are discussed.
Pediatricians encounter a relatively small number of bacterial pathogens on a frequent basis. S. aureus, Streptococcus pyogenes, S. pneumoniae, H. influenzae, Moraxella catarrhalis, and E. coli are the most common pediatrie pathogens. Klebsiella, Enterobacter, Mycoplasma, Salmonella, Shigella, Neisseria gonorrhoeae, Chlamydia trachomatis, and Chlamydia pneumoniae are also encountered in most practices, but less frequently.
Ninety-five percent of S. aureus are resistant to penicillin due to penicillinase production. But this organism is usually susceptible to semi-synthetic penicillins, cephalosporins, and clindamycin. Resistance to macrolide antimicrobials is seen in approximately 10%. Methicillin-resistant S. aureus (MRSA) is uncommon in pediatrie community-acquired infections, but is occasionally seen in institutionalized or chronically ill children. The resistance is caused by altered penicillin binding proteins, and effective oral antimicrobials are not available in this situation.
S. pyogenes remains universally susceptible to penicillin and is broadly susceptible to cephalosporins and dindamycin. Macrolide resistance has been common outside the United States when and where these agents have been used heavily. This is increasing in the United States as well. S. pyogenes infection does not respond to sulfa or trimethoprim-sulfamethoxazole (TMPSMX).
Penicillin, first-generation cephalosporins, and erythromycin do not have sufficient activity to treat H. influenzae. In most communities, 30% to 40% of H. influenzae are resistant to ampicillin and amoxicillin, due to plasmid-encoded ß-lactamase. Second-generation and third-generation cephalosporins are stable to this ß-lactamase. Azithromycin has significantly enhanced in vitro activity against H. influenzae compared with erythromycin and, like TMP-SMX, is unaffected by ß-lactamase.
Ninety percent to 100% of strains of M. catarrhalis produce ß-lactamase and therefore are resistant to amoxicillin and ampicillin. Erythromycin, the newer macrolides, secondgeneration and third-generation cephalosporins, and TMP-SMX are active against M. catarrhalis.
Drug-resistant S. pneumoniae (DRSP) are an increasing problem for pediatricians.6'7 Penicillin resistance is graded as intermediate (penicillin minimal inhibitory concentration [MIC] 0.1-1.0 /ig/ml) or high level (^ 2.0 ^g/rnl) and is due to altered penicillin binding proteins. DRSP have increased MICs to all cephalosporins, although the extent of the increase is variable for each drug. Amoxicillin-clavulanate potassium would not be expected to be more active against these isolates than amoxicillin alone. Serum levels of amoxicillin, ampicillin, or penicillin usually exceed the MIC of DRSP, but the level achieved in cerebrospinal fluid, sinuses and the middle ear. and other tissues is lower. DRSP are far more likely than fully penicillin-susceptible S. pneumoniae to be resistant to TMP-SMX (40% to 80% resistant), erythromycin and the newer macrolides (20% to 50% resistant), and clindamycin (10% to 15% resistant).
Oral Cephalosporins and Cephems Grouped by Generation
Penicillin remains the drug of choice for group A streptococcal pharyngitis, prophylaxis of acute rheumatic fever, dental infections, many oral anaerobic infections, and treatment of syphilis. Several less commonly encountered organisms are generally penicillin susceptible, such as actinomycetes, bacillus, Barretta, clostridia, leptospira, and Neisseria meningitidis.
Penicillin, ampicillin, and amoxicillin are usually equivalent for oral therapy for minor infections due to penicillin-susceptible organisms. Although amoxicillin may be preferred due to good absorption and taste, it has a broader spectrum than penicillin and thus is more likely to select resistant flora. Administration of high-dose amoxicillin or penicillin may achieve levels sufficient to overcome DRSP for soft tissue infections.
Amoxicillin-clavulanate potassium is an inhibitor of ß-Iactamase and is useful for therapy for staphylococcal infections, H. influenzae, and some gram-negative rods such as E. coli that may produce ß-lactamase. It is also useful for animal and human bites, where a mixture of aerobes and anaerobes may be encountered. Therapy is usually chosen to cover penicillin-susceptible organisms, including Pasteurella multodda from animal bites, Eikenella corrodens from human bites, other anaerobes, and S. aureus, which is often present on the skin. Amoxicillin-clavulanate potassium is not more active than amoxicillin for S. pneumoniae resistant to penicillin. However, amoxicillin given in combination with amoxicillin-clavulanate potassium may enhance the activity against DRSP with stability to the ß-lactamase produced by H. influenzae and M. catarrhalis.
Semi-synthetic penicillins, such as cloxacillin sodium and dicloxacillin sodium, are stable to the ß-lactamase produced by S. aureus. They are commonly used for minor skin and soft tissue infections where S. aureus or S. pyogenes are suspected. They are not useful for H. influenzae, E. coli, or other gram-negative organisms.
Cephalosporins and cephems are a large group of antimicrobials that are often confusing because many of the drugs Have similar properties and names (Table 1). The Cephalosporins are often grouped as first-generation, second-generation, and third-generation agents. This classification is not always precise, but it may be helpful to clinicians. Familiarity with one or two antimicrobial agents within each class is usually sufficient for office practice. Often only the least expensive drug with similar properties is available on the formulary. None of the available Cephalosporins are useful for infections caused by enterococcus, Listeria tnonocytogenes, or anaerobes.
Cephalexin is typical of the first-generation Cephalosporins. Although initially considered as a broad-spectrum antimicrobial, its range of activity against gram-negative organisms is limited. Cephalexin is active against gram-positive cocci such as S. aureus, S. pneumoniae, and S. pyogenes. It is also active against many E. coli. Cephalexin lacks substantial activity against H. influenzae and M. catarrhalis. First-generation Cephalosporins are useful for skin and soft tissue infections where S. aureus and S. pyogenes are the primary concern. It is effective for pharyngitis due to group A streptococcus, although it is not the first choice. Urinary tract infections caused by susceptible E. coli are also effectively treated by cephalexin. Cephalexin is not useful for respiratory infections such as pneumonia or otitis media unless an appropriate culture indicates a susceptible organism. Cefadroxil has activity similar to cephalexin and can be dosed twice daily, rather than the four times daily schedule used for cephalexin. Cephalexin is less expensive and is on many formularies.
Cefuroxime is a cephem antimicrobial classified as a second-generation cephalosporin. Cefuroxime has good activity against gram-positive cocci, exceeding that of some first-generation cephalosporins. Also, it has sufficient activity against gram-negative organisms such as H. influenzae and M. catarrhalis to be useful for therapy for respiratory infections and for E. coli urinary tract infections. Cefuroxime has significantly better activity against S. pneumoniae than other first-generation and second-generation cephalosporins. It also has somewhat better activity against S. aureus, and has similar activity against H. influenza and M. catarrhalis. For many years, parenteral cefuroxime was available, but there was difficulty producing an acceptable-tasting suspension. Currently, acceptable suspensions and tablets are available.
The third-generation cephalosporins have significant activity against a wide range of gramnegative organisms. Unfortunately, these drugs are generally less active than second-generation cephalosporins against gram-positive cocci. The individual drugs classified as third-generation cephalosporins differ considerably in activity against S. pneumoniae and S. aureus.
The first available oral third-generation cephalosporin was cefixime. It is less active against S. aureus than either the first-generation or the second-generation cephalosporins and is less active against S. pneumoniae than cefuroxime. As a result, cefixime is less useful for therapy for respiratory infections such as otitis media, sinusitis, and pneumonia, where S. pneumoniae is the most commonly encountered pathogen, and skin and soft tissue infections, where S. aureus is usually a concern. However, cefixime has excellent activity against the gram-negative rods encountered in urinary tract infections (except organisms such as Pseudomonas aeruginosa) and gram-negative cocci, such as N. gonorrhoeae. Some preliminary studies indicate that cefixime is equivalent to ceftriaxone sodium given parenterally for initial therapy for febrile urinary tract infections in young children.8 Cefixime has also been effective for multiply resistant Shigella species in children, who are not candidates to receive fluoroquinolones.
The spectrum and the extent of antibacterial activity of ceftibuten dihydrate is similar to cefixime, as is its potential role in clinical practice. Because of decreased efficacy against S. pneumoniae in controlled comparisons, the following statement by the manufacturer is contained in the prescribing information: "Ceftibuten should be given empirically only when adequate antimicrobial coverage against S. pneumoniae has been previously administered."9 However, several studies demonstrate that S. pneumoniae can be commonly recovered from the middle ear of children who had failed to improve with therapy presumed adequate for otitis media (ie, amoxicillin).10 As such, ceftibuten dihydrate (like cefixime) is less useful for respiratory infections, where S. pneumoniae is always a concern.
Cefprozil has greater in vitro activity than cefixime against S. aureus and S. pneumoniae. However, it is not as active against H. influenzae. Cefprozil is useful for therapy for skin and soft tissue infections and respiratory infections, and can potentially be given for urinary tract infections.
Cefpodoxime proxetil has much greater inherent activity than cefixime against S. pneumoniae and S. aureus, with only slightly decreased gram-negative activity. Cefpodoxime proxetil is indicated for 5-day therapy of otitis media. Cefpodoxime proxetil for 5 versus 10 days was compared with penicillin for 10 days as therapy for streptococcal pharyngitis, and bactériologie cure rates were 95%, 90%, and 78%, respectively.11 Similarly, cefpodoxime proxetil can be used as single-dose therapy for gonorrhea, and would be expected to be excellent therapy for urinary tract infections (except for P. aeruginosa). Cefpodoxime proxetil has sufficient activity against S. aureus that it may be given for skin and soft tissue infections. It has a pleasant initial taste, but the aftertaste is rather bitter and lingering. We have found that a number of common sweets mask the aftertaste (J. Ogle, personal observation, 1999).
Cefdinir (Parke-Davis, Inc., Morris Plains, NJ) is a newly approved third-generation cephalosporin, which is broadly active against pediatrie pathogens. Cefdinir is more active against S. aureus than cefuroxime or cefpodoxime proxetil, although less active against S. pneumoniae. Cefdinir has activity similar to cefpodoxime against gram-negative organisms such as H. influenza, E. coli, and M. catarrhalis, but is less active than cefixime. Cefdinir should be useful for therapy for respiratory infections due to penicillin-susceptible S. pneumoniae, skin and soft tissue infections due to S. aureus, and urinary tract infections.
MACROLIDES AND AZILIDES
Erythromycin is the most familiar macrolide antimicrobial and has broad activity against S. aureus, S. pyogenes, S. pneumoniae, Bordetella pertussis, C. pneumoniae, Mycoplasma pneumoniae, and Legionella pneumophuia. There are several different formulations of erythromycin, including base, stéarate, estuiate, and ethyl-succinate. The estolate gives the highest serum levels and is associated with cholestasis in adults. The base and the stéarate should be taken with meals to enhance absorption. Dose-related gastrointestinal side effects can limit the use of erythromycin. Erythromycin (and clarithromycin) have significant drug interactions with theophylline, cydoserine, carbamazepine, and terfenadine that may require dosage modifications.12
Macrolides have been useful in pediatrics because of this broad spectrum of activity. The development of new macrolides and related azilides with decreased gastrointestinal side effects, more convenient dosing, and enhanced activity has been welcome. Patients who are truly allergic to erythromycin also have allergic reactions to the newer agents. Resistance to macrolides occurs by two mechanisms: ribosomal mutation and antibiotic efflux. Normally, organisms that acquire resistance to erythromycin are also resistant to the newer macrolides and azilides.
Clarithromycin causes fewer gastrointestinal side effects compared with erythromycin and can be dosed twice daily. Clarithromycin also has a similar spectrum but is more active than erythromycin against many pediatrie pathogens. A metabolite of clarithromycin is active against H. influenzile. Clarithromycin is effective against many strains of non-tuberculous atypical mycobacteria.
Azithromycin is the first azilide antimicrobial. It has different and useful pharmacokinetic properties. Peak serum concentrations after oral dosing are low, but tissue concentrations are much higher and the half-life in tissue is prolonged. The standard dosing schedule, once daily for 5 days, is sufficient to achieve a prolonged antimicrobial effect. Azithromycin is more active than erythromycin against most pediatrie pathogens. The in vitro activity against H. influenzae is excellent, but some investigators have reported that azithromycin does not reliably eradicate H. influenzae from the middle ear.13 It is useful as single-dose therapy of chlamydia cervicitis and urethritis in sexually active teenagers. Because of activity against M. pneumoniae, C. pneumoniae, and other bacterial respiratory pathogens, azithromycin has proved useful for treatment of community-acquired pneumonia in adults, older children, and adolescents. Azithromycin absorption is affected by food and the drug should be taken 1 hour before or 2 hours after eating. Gastrointestinal complaints occur in 5% to 6% of patients, but the drug interactions common to erythromycin are reduced. Because of the low peak serum concentrations achieved with azithromycin, other drugs are preferred for children with bacteremia.
Sulfa compounds are the oldest class of antimicrobials, and many individual drugs have been formulated. Resistance to sulfonamides developed rapidly and this limits their use in infections due to S. aureus, streptococcus, S. pneumoniae, H. influenzae, and some strains of E. coli. The combination of TMP-SMX has greater activity man sulfonamide alone, but widespread use for respiratory and urinary infections has still contributed to resistance. TMP-SMX has been given for shigellosis, but resistance is increasing. Sulfa drugs cause dermatologie side effects more commonly than other antimicrobials. These include urticaria, photosensitivity, macular-papular rashes, and Stevens-Johnson syndrome. Sulfas remain useful for therapy for respiratory and urinary tract infections due to susceptible organisms.
Formulation and Route of Administration of Fluoroqulnolones*
Tetracyclines are broad-spectrum antimicrobials that are not commonly used in children because more active antimicrobials are often available. Tetracyclines are active against rickettsia, mycoplasmas, P. multocida, and a large number of other uncommon organisms. Doxycycline is preferred for therapy for C. trachomatis in pelvic inflammatory disease, cervicitis, and urethritis. Tetracyclines are commonly used for acne in adolescents. Nausea and vomiting, photosensitization, and mucosal candidiasis are common side effects. Tetracyclines may stain the permanent teeth in children younger than 9 years, but a single course of therapy does not pose a significant risk.
Clindamycin is active against gram-positive cocci such as S. aureus, including MRSA, S. pyogenes, most strains of S. pneumoniae, other streptococcal species except enterococcus, and both gram-positive and gram-negative anaerobes.14 It is also active against C. trachomatis and has in vitro activity (but not proven efficacy) against N. gonorrhoeae. Clindamycin inhibits protein synthesis and has been used for toxin-mediated grampositive infections, such as the toxic shock syndrome due to S. aureus or S. pyogenes. Clindamycin is 90% absorbed orally, and food delays but does not decrease absorption. It achieves high concentrations in oral secretions, and is useful for dental and other intra-oral infections, the eradication of group A streptococcal carriage state, and oral therapy for bone infections. Diarrhea and gastrointestinal side effects are common. Fseudomembraneous colitis due to Clostndium difficile was originally associated with clindamycin but is now known to occur following most antimicrobials.
Metronidazole has activity against anaerobic organisms, gram-negative anaerobic organisms, and many protozoan species.14 It is commonly used for giardiasis, trichomonads, amebiasis, and bacterial vaginosis. Metronidazole is well absorbed orally. The taste is bitter with a prolonged aftertaste, and this often causes young children to refuse to take the drug. Diarrhea and vomiting are the most common side effects. Rare serious neurologic side effects, including seizures, are reported. Concerns have been raised about the mutagenicity and carcinogenicity of metronidazole because of experiments in rats and mice. Long-term follow-up studies of women treated with metronidazole have not shown significant overall increased risk of cancer. Slightly increased rates of lung cancer were confounded by smoking among women.14
Nalidixic acid, an old antimicrobial that is not currently useful, was the initial drug in this class. Modification of the basic quinolone structure resulted in the fluoroquinolones, which possess broad antibacterial activity, excellent oral bioavailability, and favorable side effect profiles.15 Ten quinolones are available in the United States, including intravenous, otic, and ophthalmic preparations, tablets, and suspensions (Table 2). Similar to what has happened with the cephalosporins, there has been a profusion of new drugs with similar-sounding names and properties. These agents are inhibitors of DNA gyrase, and therefore are not affected by ß-lactamases that inactivate penicillins and cephalosporins. Floroquinolones are active against gram-negative organisms, including otherwise resistant E. coli, Klebsiella, Enterobacter, Salmonella, Shigella, Campylobacter, Haemophilus, and Moraxella. Several of the drugs possess excellent activity against P. aeruginosa. Some are useful for the sexually transmitted organisms N. gonorrhea and C. trachomatis (eg, ofloxacin, trovofloxacin, and grepefloxacin), as well as mycoplasma, C. pneumoniae, N. meningitidis, and legionella. Although the initial fluoroquinolones (eg, norfloxacin or ciprofloxacin) had limited activity against S. aureus and S. pneumoniae, the newer drugs are active against S. pneumoniae, including DRSP, and anaerobes.
Quinolone antimicrobials (except the otic and ophthalmic preparations) are not approved in children. Nalidixic acid and, subsequently, other quinolones were shown to injure the cartilage of actively growing animals. However, arthropathy has not been a problem in adults nor in the many children, primarily with cystic fibrosis, given fluoroquinolones under compassionate release programs. Long-term follow-up of children who received nalidixic acid has shown normal growth and joint function without arthritis. Some pediatrie providers are using fluoroquinolones when clear advantages exist over current alternatives. The fluoroquinolones are usually well tolerated, although gastrointestinal complaints, insomnia and minor central nervous symptoms, and photosensitization are side effects.16
Pediatrie providers should use fluoroquinolones only if the advantages clearly putweigh the risks and if alternative therapy is not available or is disadvantageous.17 Parents should be informed of risks and advantages and that the drug being prescribed is not approved in children.
RECOMMENDATIONS FOR THERAPY
Strict consensus recommendations for therapy for common pediatrie infections do not exist. Significant geographic differences are found for prescribing practices and resistance rates. Sometimes these are even present within a single metropolitan area. The antimicrobial agent of choice is influenced in part by availability on the formulary, insurance coverage and cost, and parental preferences for convenient dosing schedules, palatability, and side effect profiles. For most pediatrie infections, several antimicrobials of equal efficacy are available. Practitioners should always favor older drugs with known side effect profiles, and those that have a narrow spectrum versus a broad spectrum.
ACUTE OTITIS MEDIA (AOM)
AOM is primarily treated to reduce the incidence of mastoiditis, perforation of the tympanic membrane, and other suppurative complications. Prevention of hearing loss is an important secondary goal. It is estimated that without antimicrobial therapy, two-thirds to three-fourths of AOM cases would resolve without sequelae. However, guidelines to identify patients unlikely to have complications of AOM have not been validated. Therefore, AOM should be treated to improve symptoms in those whose conditions would not spontaneously resolve and to prevent suppurative complications. S. pneumoniae, H. influenzae, and M. catarrhalis are the most common pathogens. Viruses almost certainly cause AOM, but the clinician cannot reliably separate viral and bacterial disease. The diagnosis of AOM may be difficult, and many different criteria are used to establish the diagnosis.18 The likelihood of antimicrobial resistance is greater in children who have recently received an antimicrobial and those in day care.
Amoxicillin (40-45 mg/kg/d divided in three doses) is the first choice for initial therapy because of cost effectiveness, and side effect profile. In communities where DRSP are common, higher dose therapy (80-90 mg/kg/d divided in three doses) is more likely to achieve middle ear concentrations sufficient to inhibit these organisms. Response to therapy should then be monitored during the successive 3 days. Children with persistent fever, pain, bulging of the tympanic membrane, or otorrhea may be treatment failures and alternative therapy should be considered. Such second-line therapy should be effective against both DRSP and ß-lactamase-producing H. influenzae and M. catarrhalis.
An ideal second-line oral drug is not available. The azilides and macrolides (eg, azithromycin and clarithromycin) may fail to eradicate H. influenzae from the middle ear and resistance is common among DRSP. The cephalosporins are better than amoxicillin for ß-lactamase-producing organisms, but not for DRSP. Of the oral cephalosporins, cefpodoxime proxetil and cefuroxime are most active against DRSP, but have a narrow therapeutic ratio in the middle ear. TMP-SMX and erythromycin-sulfamethoxazole cover H. influenzae and M. catarrhalis, but, again, DRSP are commonly resistant. Clindamycin has excellent S. pneumoniae activity, but little effectiveness against gram-negative organisms. A standard dose of amoxicillin-clavulanate potassium, combined with amoxicillin to achieve a total daily amoxicillin dose of 80 to 90 mg/kg, will cover ß-lactamase-producing H. influenzae and M. catarrhalis. But this combination does not improve coverage against DRSP. Ceftriaxone, 50 mg/kg intramuscularly for 3 days, should achieve high middle ear concentrations sufficient for resistant pathogens, but requires parenteral therapy and additional expense.19 The author prefers amoxicillin-clavulanate potassium (dosed at 80-90 mg/kg/d of amoxicillin), TMP-SMX, cefuroxime, or cefpodoxime proxetil for such second-line therapy and reserves intramuscular ceftriaxone as third-line treatment. When available, rympanocentesis should be used to define microbiology (the agent and its sensitivities) in treatment failures.
Children with AOM in the United States have usually been treated for 10 days, although shorter duration therapy has been standard in other countries. More recently, 5 to 7 days has been recommended for low-risk children. Age younger than 24 months, perforation of the tympanic membrane, and perhaps prior frequent otitis media are suggested criteria for the exclusion of shorter duration therapy. However, data to establish these criteria are incomplete.18
The goal of therapy is to prevent complications and reduce unnecessary treatment while minimizing expense. Pediatricians can more easily direct therapy if patients are treated in the office rather than by decisions made after hours in the emergency department. It may require considerable work to teach families that otitis can be treated the following day unless unusually severe. In the office, the practitioner should strive to minimize therapy for otitis media with effusion and persistent middle ear effusion following otitis, and to eliminate therapy for uninfected but scarred or otherwise abnormal-appearing tympanic membranes. It is difficult for an emergency medicine physician, confronted by a febrile, irritable child with an abnormal tympanic membrane and a recent history of otitis media, to distinguish a new infection from persistent middle ear fluid remaining after adequate treatment.
Prophylaxis of frequently recurrent AOM has been shown efficacious in several trials. But the benefit of prophylaxis must be weighed against the likelihood of selection of resistant strains.18 If prophylaxis is used, sulfisoxazole or amoxicillin is most appropriate and prophylaxis should be given for no more than 6 months and limited to children with documented frequent recurrences. Other steps to reduce the incidence of AOM, such as reducing smoking in the home, reducing day care attendance, and administering influenza vaccine, may be beneficial. However, each of these may be difficult or impractical.
Sinusitis, the accumulation of fluid with bacterial overgrowth in the paranasal sinuses, is a common consequence of viral upper respiratory tract infection and allergy. Most cases of sinusitis will resolve without specific therapy.20 The ethmoid and maxillary sinuses are rudimentarily present at birth, but the frontal and ethmoid sinuses are in this state until 6 years of age. So clinically significant frontal or sphenoid sinusitis is unusual prior to adolescence.
Sinusitis is a clinical diagnosis. Sinus radiographs are often abnormal in children who do not need treatment, especially those younger than 1 year of age, and after upper respiratory tract infections at any age, including in adults. Sinus radiographs or computerized tomography scans should thus be reserved for children with recurrent sinusitis or for those with complications or in need of surgery. Prospective evaluations of children with otherwise uncomplicated colds that persist for 10 to 30 days established the benefit of antimicrobial therapy. However, by this criterion, approximately five children must be treated to benefit one.21
On the other hand, some children may present with signs of severe acute sinusitis. They may have high fever (temperature > 390C), facial tenderness or swelling, fetid odor to their breath, or dental-type pain. These children should be considered for inpatient parenteral therapy due to the high incidence of complications.
Few studies have defined the microbiology and response to therapy in children with acute sinusitis. The studies available suggest that the organisms recovered are the same as those seen in AOM. The antimicrobial choices and duration of therapy appropriate for AOM should be used for sinusitis. Data to establish the duration of therapy for acute sinusitis are not available. To minimize the duration of unnecessary antimicrobials, some authors suggest therapy for 5 days beyond the resolution of clinical symptoms. This may sound like a practical solution, but could still result in excessively prolonged therapy for children with atypical or mild symptoms.
Pharyngitis is most commonly caused by viral infections in children of all ages. Differentiating bacterial infection caused by S. pyogenes from a viral infection requires a culture or an appropriately performed rapid test for streptococci. Therapy with penicillin within 9 days after onset of symptoms substantially reduces the incidence of acute rheumatic fever. Penicillin, amoxicillin, or ampicillin is the drug of choice for streptococcal pharyngitis. Many studies have demonstrated that bactériologie and clinical cure rates for streptococcal pharyngitis are slightly greater for cephalosporins than for penicillin. Group A ßhemolytic streptococci that are resistant to penicillin have not been described. Postulated reasons for the apparent superiority of cephalosporins include ß-Iactamase production by pharyngeal staphylococci or anaerobes that protect group A streptococci from penicillin, and increased activity of cephalosporins against carriers.22
Nonetheless, penicillin remains the drug of choice because of cost, narrowness of spectrum, absence of side effects, and lack of selection of resistant flora. Macrolides and azilides are alternatives for children who are allergic to penicillin. All of the cephalosporins are effective and have been used as therapy for symptomatic, culturepositive treatment failures. Twice and three times daily dosing schedules are convenient and have established efficaciousness for most antimicrobials. Cefpodoxime proxetil and azithromycin were effective when given for 5 days. A recent study also showed the efficacy of 750 mg of amoxicillin once daily.
There is no accepted definition of streptococcal carriers. Neither antibody studies nor cultures unambiguously define children who are carriers. Nonetheless, many practitioners see children with repeatedly positive throat cultures in the absence of clinical disease. These children are probably streptococcal carriers. Clindamcyin or penicillin given with rifampin is effective in clearing streptococcal carriage, when treatment is deemed necessary.
Lower respiratory tract infections in children are most commonly viral in etiology. Differentiation of viral and bacterial disease can be difficult because laboratory parameters and clinical symptoms may overlap. The presence of pneumatocele, lung abscess, and empyema strongly suggest bacterial disease. Similarly, focal infiltration, toxicity, and elevated temperature and white count suggest bacterial rather than viral infection. Pediatrie providers are often faced with the difficult task of differentiating bacterial and viral lower respiratory tract infections.
The bacteria most commonly associated with lower respiratory tract infection in infants and young children are the same as those that cause AOM and sinusitis - S. pneumoniae, H. influenzile, and M. catarrhalis. Because S, pneumoniae is the most common cause of bacterial pneumonia, amoxicillin is usually the drug of choice. TMP-SMX, amoxicillin-clavulanate potassium, a second-generation or third-generation cephalosporin, clarithromycin, or azithromycin is also efficacious. In the absence of intrapulmonary complications such as empyema, there is no evidence that a cephalosporin, a macrolide or azilide, or amoxicillin-clavulanate potassium has greater activity against the common pathogens (or a better clinical outcome) than penicillin. In older children, M. pneumoniae and C. pneumoniae are more common, and a macrolide or azilide will cover common bacteria and these atypical organisms. Similarly, children younger than 4 months with afebrile pneumonia are often infected by C. trachomatis and thus will benefit from a macrolide.
URINARY TRACT INFECTION
Pediatrie practitioners need to be more vigilant in seeking evidence of urinary tract infections because overdiagnosis is common and failure to diagnose this may lead to renal scarring and complications due to renal injury. Proper urine collection is difficult in children. Bag and clean-catch urine specimens collected from infants and children are frequently contaminated, even if a single species in high concentration is recovered. Suprapubic specimens or urine specimens collected by catheter are less likely to be contaminated and provide a more secure basis for diagnosis.
E. coli is the cause of more than 90% of urinary tract infections in children. Both the diagnosis of urinary tract infection and the localization to the upper versus the lower urinary tract are difficult in young children. Neither fever, systemic signs, white blood cell count, nor other laboratory tests are accurate enough to reliably differentiate upper from lower tract infections.
Sulfas, penicillins, and cephalosporins are concentrated in the lower urinary tract, providing a therapeutic advantage for treatment of lower tract disease. Uncomplicated urinary tract infection in mildly ill children will probably respond to amoxicillin, sulfa, TMP-SMX, or cephalosporins. Tetracycline or a fluoroquinolone may be used in adolescents. Nitrofurantoin and nalidixic acid should not be used for pyelonephritis or in young children, for whom the differentiation of upper and lower tract disease is difficult. Severe and complicated infections, especially in children with abnormal urinary systems, will respond better to parenteral therapy with a cephalosporin or aminoglycoside (eg, ceftriaxone or amikacin). Children who are toxic, vomiting, or dehydrated; children in whom follow-up may be unreliable; or children with abnormal urinary tracts should be considered for parenteral therapy in the hospital.
SKIN AND SOFT TISSUE INFECTIONS
S. aureus and S. pyogenes are most commonly the causes of skin and soft tissue infections. Many choices for antimicrobial therapy are available. First-generation and second-generation cephalosporins, such as cephalexin, cefadroxil, cephradine, and cefuroxime, and semi-synthetic penicillins, such as dicloxacillin sodium, have good activity and equal efficacy. The third-generation cephalosporins - cefprozil, cefpodoxime, and Cefdinir - have good anti-staphylococcal and anti-streptococcal activity. Macrolide and azilide antimicrobials also have good activity, but 10% of isolates are resistant in many geographic locations. The cost of prescriptions should be an important factor in the decision for therapy.
The selection of appropriate outpatient antimicrobials for children will continue to evolve as new data and drugs become available. The primary care provider must identify children who require an antimicrobial, and educate families that antimicrobials are often unnecessary. Whenever practical, narrow spectrum, older, less expensive drugs are preferred.
1. McCraig L, Hughes J. Trends in antimicrobial drug prescribing among office-based physicians in the United States. JAMA. 1995;273:214-219.
2. Mainous AG ??, Hueston WJ, Clark JR. Antibiotics and upper respiratory tract infection: do some folks think there is a cure for the common cold? / Fam Pract. 1996;42:357-361.
3. Schwartz R, Frei) B, Ziai M, Sheridan M. Antimicrobial prescribing for acute purulent rhinitis in children. Pediatr Infect Dis J. 1997;16:185-190.
4. Dowell SF, Marcy SM, Phillips WR, Gerber MA, Schwartz B. Principles of judicious use of antimicrobial agents for pediatrie upper respiratory tract infection. Pediatrics. 1998;! 01: 163-164.
5. Bauchner H, Pelton SI, Klein JO. Parents, physicians, and antibiotic use. Pediatrics. 1999;103:395-398.
6. Dowell S, Butler JC, Giebink GS, et al. Acute otitis media: management and surveillance in an era of pneumococcal resistance: a report from the drug-resistant Streptococcus pneumoniae working group. Pediatr Infect Dis }. 1999;18:l-9.
7. Fung-Tomc JC, Huczko E, Stickle T, et al. Antibacterial activities of cefprozil compared with those of 13 oral cephems and 3 macrolides. Antimicrob Agents Chemother. 1995;39:533-538.
8. Hoberman A, WaId ER. Urinary tract infections in young febrile children. Pediatr Infect Dis }. 1997;16:11-17.
9. Physician's Desk Reference. Montvale, NJ: Medical Economics; 1999:2820.
10. Leibovitz E, Raiz S, Piglansky L, et al. Resistance pattern of middle ear fluid isolates in acute otitis media recently treated with antibiotics. Pediatr Infect Dis }. 1998;17:463469.
11. Pichichero ME, Gooch WM, Rodríguez W, et al. Effective short-course treatment of acute group A ß-hemolytic streptococcal tonsUlopharyngitis. Arch Pediatr Adúlese Med. 1994;148:1053-1060.
12. Reed MD, Blumer JL. Azithromycin: a critical review of the first azilide antibiotic and its role in pediatrie practice. Pediatr Infect Dis]. 1997;18: 1069-1083.
13. Klein JO. The "in vivo sensitivity test" for acute otitis media. Pediatr Infect Dis J. 1998;! 7:774-775.
14. Falagas ME, Gorbach SL. Clindamycin and metronidazole. Med Clin North Am. 1995;79:845-866.
15. Suh B, Lorber B. Quinolones. Med Clin North Am. 1995;79:869-893.
16. Lipsky BA, Baker CA. Fluoroquinolone toxicity profiles: a review focusing on newer agents. Clin Infect Dis. 1999;28:352-364.
17. Schaad UB, Salam MA, Aujard Y, et al. Use of fluoroquinolones in pediatrics: consensus report of an International Society of Chemotherapy commission. Pediatr Infect Dis }. 1995;14:l-9.
18. Dowell SF. Otitis media: principles of judicious use of antimicrobial agents. Pediatrics. 1998;101:165-171.
19. Leibovitz E, Piglanski L, Raiz S, et al. Bactériologie efficacy of a three day intramuscular ceftriaxone regimen in nonresponsive acute otitis media. Pediatr Infect Dis J. 1998;17:1126-1131.
20. O'Brien K, Dowell SF, Schwartz B, Marcy SM, Philips WR, Gerber MA. Acute sinusitis: principles of judicious use of antimicrobial agents. Pediatrics. 1998;! 01:1 74-177.
21. WaId ER, Chiponis D, Ledesma-Medina J. Comparative effectiveness of amoxicillin and amoxicillin-clavulanate potassium in acute paranasal sinus infections in children: a double-blind, placebo-controlled trial. Pediatrics. 1986;77:795-800.
22. Schwartz B, Marcy SM, Phillips WR, Gerber MA, Dowell SF. Pharyngitis: principles of judicious use of antimicrobial agents. Pediatrics. 1998;! 01:1 71-1 74.
Oral Cephalosporins and Cephems Grouped by Generation
Formulation and Route of Administration of Fluoroqulnolones*