A Clinical guide to pediatric infectious disease


Infection in Unusual Spaces

Certain pediatric infections are noteworthy for their presence in unusual body spaces. These may be rarely seen but when present require rapid diagnosis and therapy. This chapter provides a discussion of such infections, including necrotizing fasciitis, omphalitis, endophthalmitis, peritonitis secondary to ruptured appendicitis, and retropharyngeal abscess.

Necrotizing Fasciitis

Necrotizing fasciitis is an infection involving the subcutaneous tissues and deep fascia. It can affect any portion of the body, although the lower extremities are the most commonly involved. First described by Hippocrates in the 5th century, this condition has been reported under a variety of names, including “hospital gangrene” and “malignant ulcer.” The term necrotizing fasciitis was first used in the 1950s and accurately describes the location of infection.


The infection begins with the introduction of the pathogen into the subcutaneous tissues. Many mechanisms of this introduction have been reported, including insect bites, minor trauma, preceding varicella infection, and surgical incisions. Hematogenous spread has also been reported as a means of inoculation. A variety of toxins, cytokines, and inflammatory mediators are thought to be involved in the progression of the infection.

Necrotizing fasciitis has been divided into distinct groups based on causative organism. Type 1 refers to a polymicrobial infection usually caused by non–group A streptococcus and other aerobic and anaerobic bacteria. Type II necrotizing fasciitis usually is caused by group A streptococcus alone or with staphylococcus. The etiologic agents of necrotizing fasciitis cannot be determined from clinical presentation alone. During the past decade, the most common cause of necrotizing fasciitis has remained group A streptococcus following varicella infection. It is thought that the group A streptococcus is inoculated directly into the skin when the child scratches the varicella lesions.



A major challenge for the pediatrician is to separate necrotizing fasciitis from a routine cellulitis. Clinical examination is the mainstay of diagnosis. A major clue on physical examination involves severe pain, often out of proportion to the physical findings. As the infection progresses, one finds worsening erythema and edema. Later, the skin may develop blisters, and bullae may form (Fig. 16.1). The formation of bullae is thought to be an important diagnostic finding that should always raise the suspicion for necrotizing fasciitis. Later in the progression of the disease, the bullae become hemorrhagic and are often accompanied by crepitus.


Several tests have been used in an attempt to document the progression of cellulitis to necrotizing fasciitis. The presence of leukocytosis and acidosis has been used to identify patients with progressive disease. The appearance of gas on plain radiograph is an inconsistent finding and is seen in less than 20% of cases. Magnetic resonance imaging (MRI) has been reported to detect extension of the inflammatory process into the subcutaneous tissue. High signal intensity of the fascia in the T2-weighted images strongly suggests the diagnosis.


FIG. 16.1. Cellulitis, edema, and bullae formation in child with necrotizing fasciitis (see color plate).



The gold standard of diagnosis for necrotizing fasciitis is surgical biopsy; biopsies reveal acute inflammation of the dermis and fascia with accompanying thrombosis of blood vessels.


The management of necrotizing fasciitis involves aggressive medical and surgical therapy. Optimal medical treatment includes a third-generation cephalosporin and anaerobic coverage, usually with clindamycin. Complete débridement of all devitalized tissue is required. A repeat second-look surgery after 24 hours is often needed to determine whether remaining devitalized tissue is present. Adjunctive therapies include hyperbaric oxygen and intravenous immunoglobulin, although definitive data of the efficacy of these measures is lacking (Table 16.1).


Omphalitis, or infection of the umbilicus, remains a major cause of morbidity and mortality in developing countries. In a poorly understood process, omphalitis is thought to arise from bacterial colonization occurring at the time of delivery. This colonization becomes invasive and can proceed to funisitis (a term given to mild cellulitis or inflammation of the periumbilical skin) or to omphalitis, in which the umbilical stump and surrounding tissues are involved.


Omphalitis has been traditionally caused by Staphylococcus aureus and Streptococcus pyogenes (group A streptococcus.) Modern cord care with triple-dye antimicrobial soap and alcohol has reduced the frequency of omphalitis in developed countries; however, it may also have changed the microbiology of omphalitis. During


the past 10 years, reports of omphalitis have emphasized the role of additional pathogens, particularly gram-negative and anaerobic bacteria.

TABLE 16.1. Management of Necrotizing Fasciitis


1. Pain out of proportion to physical exam

2. Induration, “woody” edema

3. Leukocytosis, acidosis

4. Fascial involvement on MRI

Medical treatment

1. Third generation cephalosporin; cefotaxime 50 mg/kg IV q8

2. Anaerobic coverage; Clindamycin 20–40 mg/kg/d divided q8

Surgical treatment

1. Complete debridement of devitalized tissue

2. “Second look” surgery may be needed

Adjunctive care

1. Hyperbaric oxygen

2. Intravenous immunoglobulin (400 mg/kg/dose)


Neonates typically present with purulent discharge of the umbilical cord with rapidly progressing cellulitis of the abdominal wall.


The diagnosis is based largely on the history and clinical examination. It is generally thought that any abdominal wall cellulitis surrounding the umbilical structures is consistent with the diagnosis of omphalitis.


Due to the potential life-threatening complications of an umbilical cord infection, neonates presenting with any evidence of inflammation around the umbilical cord should be managed aggressively. This includes admission and treatment with broad-spectrum antibiotics, usually a third-generation cephalosporin and clindamycin. Surface cultures should be obtained, although it should be stressed that these may not represent the entire spectrum of bacteria involved in the deeper fascial planes. As in necrotizing fasciitis, surgical resection is a major part of therapy, and early involvement with experienced pediatric surgeons is mandatory.

Continued concern regarding omphalitis has led to an ongoing evaluation of optimal umbilical cord care. Some hospitals have abandoned triple-dye alcohol regimens for a regimen of dry cord care consisting of gentle cleaning with soap and water and allowing the area to air dry. A prospective study evaluating this approach in nearly 800 infants found a single case of omphalitis in the dry care group. In the dry care group, infants were more likely to be colonized with group B streptococcus, S. aureus, and gram-negative bacteria. Whether this colonization ultimately leads to an increase in invasive infection will need to be the subject of further studies (Table 16.2).

TABLE 16-2. Management of Omphalitis

Medical therapy

1. Polymicrobial infection including S. aureus, group A streptococcus, anaerobes

2. Surface cultures may not reflect all causative pathogens

3. Cefotaxime – 100–200 mg/kg/day in 3 divided doses

4. Clindamycin – 20–40 mg/kg/day in 3 divided doses

Surgical therapy

1. Complete excision of affected tissue




Endophthalmitis refers to infection within the ocular structures. There are two mechanisms for this infection. In endogenous (hematogenous) endophthalmitis, bacteria are seeded within the eye following a bacteremic or septicemic illness. Exogenous endophthalmitis refers to infection within the eye following direct inoculation from a surgical procedure or traumatic event.


Organisms that can seed the ocular structures following a bacteremic or septicemic illness include Bacillus cereus, Candida species, andNeisseria meningitidisB. cereus infection is associated with the use of intravenous drugs. Organisms of exogenous disease includeStaphylococcus epidermidis, Streptococcus species, and S. aureus.


Patients have decreased vision and proptosis, often accompanied by periocular inflammation and edema. Visual acuity is usually markedly decreased. This process should be suspected in any bacteremic patient who, during the course of illness, develops ocular complaints. Exogenous endophthalmitis is particularly important in pediatrics because ocular trauma may not be immediately reported.


The diagnosis of endophthalmitis should be done in conjunction with an experienced ophthalmologist. Ophthalmologic examination may reveal corneal haziness and a purulent exudate (hypopyon) in the anterior chamber of the eye. Typically, anterior chamber and vitreous aspiration should be performed in an effort to identify the responsible pathogen. The yield of Gram stain or culture approaches 60%.


Therapy is difficult, given the low rate of positive cultures and the poor penetration of systemic antibiotics into the eye. Initial antibiotic choices are often based on the most likely pathogen. It is recommended that, following the aspiration of fluid, antibiotics be instilled directly into the vitreous cavity. Intraocular vancomycin, 1 mg, and ceftazidime, 2.25 mg, are frequently used. Amikacin, 400 µg, is used in some centers as an alternative to ceftazidime for gram-negative coverage.

Intravenous antibiotics are considered to have poor penetration into the aqueous humour. Adjunctive therapy with intravenous vancomycin, ceftazidime, and amikacin is frequently employed, although their usefulness remains unclear.


Ciprofloxacin has been reported to achieve levels above the minimal inhibitory concentration (MIC) for coagulase-negative staphylococcus; some centers use this medication if this pathogen is identified. Intraocular and systemic steroids, in conjunction with vitrectomy, have been used for progressive disease (Table 16.3).

TABLE 16-3. Endophthalmitis

Endogenous (septicemic)

1. Bacillus cereus, candida species, Neisseria meningitidis

Exogenous (direct inoculation)

1. S. epidermidis, Streptococcus species


1. Examination by ophthalmologist

2. Anterior chamber/vitreous aspiration

3. Positive gram stain, culture in 60%


1. Intraocular medications
   Vancomycin 1 mg
   Ceftazidine 2.25 mg
   Amikacin 400 µg

2. Vitrectomy

Peritonitis Secondary to Ruptured Appendicitis


Children with appendicitis often have perforation at the time of diagnosis. This perforation leads to the seeding of the peritoneal cavity with the multitude of aerobic and anaerobic organisms found in the gastrointestinal tract. This seeding serves as a risk factor for the development of intra-abdominal abscesses. During recent years, there has been a great interest in the proper management of patients with peritonitis following a perforated appendicitis.


Although adults frequently present with periumbilical pain and subsequent migration of point tenderness to the right lower quadrant, up to one half of young children present with the appendicitis already ruptured. These children often present with fever, diffuse abdominal tenderness, and rebound tenderness indicating peritoneal irritation.


Although the hallmark of infectious disease evaluation is obtaining appropriate cultures, there remains debate about the clinical usefulness of obtaining intraoperative peritoneal cultures in patients with perforated appendicitis. Some studies have


demonstrated that culture results do not alter therapy or outcome in patients with peritonitis. This appears to be particularly true whenPseudomonas aeruginosa is isolated; in this patient population, P. aeruginosa is not considered a clinically important pathogen. Other authors cite alternative studies in which complications were more common in patients whose intraoperative cultures grew an organism that was not covered by the initial antibiotic regimen. One explanation for these discrepancies is that peritoneal cultures are often obtained in children with perforated but not abscessed appendicitis. The potential for abscess formation is extremely low in this population.


In 2003, the Infectious Disease Society of American (IDSA) published guidelines for the selection of antibiotics for complicated intra-abdominal infections. The study reported the clinical usefulness of cultures, particularly in complicated (i.e., those processes associated with peritonitis or abscess formation) intra-abdominal infections. The IDSA recommended that culture and susceptibility be done against the isolated gram-negative bacilli because there is increasing resistance among these organisms. Based on published reports of increasing resistance in the Bacteroides fragilis group of organisms, it recommended that empiric use of clindamycin, cefoxitin, and quinolones for treatment of B. fragilis not be used.

The polymicrobial nature of intra-abdominal infections requires consideration of a broad-spectrum regimen. Coverage against any enteric gram-negative organisms, such as Escherichia coli, and anaerobes, such as Bacteroides species, is crucial. For this reason, ampicillin, gentamycin, and clindamycin, the traditional “surgical triples,” were often used in the past for intra-abdominal infections. As a result of increasing resistance among the gram-negative enteric bacteria and B. fragilis, newer regimens are increasingly used. A single combination agent such as ampicillin-sulbactam (Unasyn) or piperacillin-tazobactam (Zosyn), is often used to treat community-acquired intraabdominal infections of mild to moderate severity. Carbapenems, which include imipenem and meropenem, can also be used as single agents because these drugs provide broad-spectrum gram-negative and anaerobic coverage. Combination regimens based on β-lactam antibiotics include third- or fourth-generation cephalosporins such as cefotaxime, ceftriaxone, and cefepime, combined with metronidazole (Flagyl). The latter provides good coverage against Bacteroides species, which may be resistant to clindamycin.

Enterococcus is a common organism found in the intestinal tract. In the past, the regimens used to treat complicated intraabdominal infections have provided coverage for this organism. A review of previous studies done by the IDSA found that coverage against enterococci does not provide an advantage and is not necessary in treatment of community-acquired intra-abdominal infection. Similar recommendations can be found for Candida albicans, which is isolated in a good percentage of patients who have perforation of the gastrointestinal tract. Unless the patient is receiving


immunosuppressive therapy or has recurrent disease, antifungal therapy is not required.

TABLE 16-4. Peritonitis Secondary to Ruptured Appendicitis

1. Organisms: Polymicrobial including enterococcus, enteric gram negatives (E. coli) and anaerobes (can no longer assume clindamycin susceptibility of Bacteroides fragilis species).

2. Antibiotics for community acquired complicated intra-abdominal infections.

1. Ampicillin/sulbactam (Unasyn): 100–200 mg/kg/day of ampicillin component divided every six hours.

2. Piperacilin/ tazobactam (Zosyn): > 6 months; 300–400 mg/kg/day of piperacillin component divided q 6–8 hours.

3. Imipenem/ cilastin: 15–25 mg/kg every six hours.

4. Combination regimens: Third or fourth generation cephalosporin: Cefotaxime (100mg/kg/day) in Ceftriaxone (75–100mg/kg/d) or Cefepime (50mg/kg q 8) combined with metronidazole (Flagyl) 30mg/kg/day in 4 divided doses.

3. Children with non-complicated disease, discontinue antibiotics when afebrile and white blood cell counts reveals less than 3% band forms, may also consider changing to oral antibiotics when tolerated to complete ten days therapy.

Antibiotic Duration

The duration of antibiotic treatment and the ability to transition to oral antibiotics have also been studied. These studies usually reflect the child with noncomplicated appendicitis in whom there is no extensive infection or abscess in the peritoneal cavity; these are the children most frequently seen by pediatricians. One recent study observed that intravenous antibiotics can be discontinued when the patient is afebrile for 24 hours and the white blood cell count shows less than 3% band forms. Typically, the duration of therapy following this protocol was between 3 and 14 days. Other investigators have evaluated programs in which patients are transitioned to oral amoxicillin-clavulanate (Augmentin) plus metronidazole to complete a 10-day total course. In the patients studied, there was no difference between children who had received oral antibiotic therapy and those who had prolonged intravenous treatment (Table 16.4).

Retropharyngeal Abscess


Retropharyngeal abscess is a deep neck infection that can obstruct the upper airway. Most cases occur in patients younger than 5 years of age. These infections are the result of bacteria in the nasopharynx or middle ear infecting the lymph nodes that lie between the posterior pharyngeal wall and the prevertebral fascia. These nodes are thought to atrophy during childhood, which may explain the age distribution seen. Traumatic retropharyngeal abscess is the result of penetrating trauma and can occur at any time in life.




Microbiology of retropharyngeal abscess has been the focus of considerable study. Group A streptococcus is the usual pathogen, although numerous investigators believe there may be a large polymicrobial component. Organisms that have been isolated in children with retropharyngeal abscess include S. aureus and anaerobic species such as Bacteroides and Peptostreptococcus species.


Patients often present with a history of sore throat which progresses to increasing neck swelling, drooling, or dysphagia. Affected children may also exhibit decreased neck mobility to such an extent that meningitis is frequently considered and lumbar puncture performed. Increasing stridor has been reported as a classic clinical sign of retropharyngeal abscess, although recent reviews have not found this to be a consistent symptom.


Diagnosis of retropharyngeal abscess begins with appropriate clinical suspicion in the right clinical setting. Lateral neck films can show an increased retropharyngeal soft tissue space, although interpretation of these plain films may be dependent on patient positioning. Computed tomography of the neck can be extremely helpful in documenting infection in the retropharyngeal space (Fig. 16.2). Evaluation by computed tomography is not without difficulty. There are limitations in the ability of computed tomography to determine definitively whether a well-defined abscess is present; a retropharyngeal cellulitis (or phlegmon) may be difficult to distinguish on computed tomography from a frank abscess. Numerous reports have stated that clinical correlation should be used in making decisions regarding the presence of a true abscess.


Treatment of retropharyngeal cellulitis or abscess always involves the use of appropriate antimicrobials. Ampicillin-sulbactam (Unasyn) offers an advantage in that it has broad-spectrum coverage against not only group A streptococcus and S. aureus but also a variety of anaerobic organisms. A combination of clindamycin and a third-generation cephalosporin will also provide coverage for the probable organisms involved. The adjunctive surgical management of retropharyngeal abscess remains controversial. Some reviews and textbooks state that the treatment always includes drainage; many otolaryngologists feel that a trial of antibiotics can be used initially. Failure to respond clinically to medical management alone warrants surgical therapy (Table 16.5).


FIG. 16.2. Computed tomography scan revealing retropharyngeal abscess.

TABLE 16-5. Retropharyngeal Abscess

1. Clinical presentation: Fever

2. Pathogens: Polymicrobial, including group A streptococci, and oral anaerobes

3. Diagnosis: Computed tomography of neck

4. Antibiotics Clindamycin 20–40 mg/kg/day in 3 divided doses in combination with a third generation cephalosporin (cefotaxime or ceftriaxone) or unasyn (ampicillin-sulbactam): 100–200 mg/kg/day of ampicillin component in 4 divided doses

5. Surgery





Selected Readings

Barton LL, Jeck DT, Vaidya VU. Necrotizing fasciitis in children: report of two cases and review of the literature. Arch Pediatr Adolesc Med1996;150(1):105–108.

Brook I. Microbiology of retropharyngeal abscesses in children. Am J Dis Child 1987;141(2):202–204.

Craig FW, Schunk JE. Retropharyngeal abscess in children: clinical presentation, utility of imaging, and current management. Pediatrics2003;111(6 Pt 1):1394–1398.

Gollin G, Abarbanell A, Moores D. Oral antibiotics in the management of perforated appendicitis in children. Am Surg 2002;68(12):1072–1074.

Hoelzer DJ, Zabel DD, Zern JT. Determining duration of antibiotic use in children with complicated appendicitis. Pediatr Infect Dis J1999;18(11):979–982.

Hsieh WS, Yang PH, Chao HC, et al. Neonatal necrotizing fasciitis: a report of three cases and review of the literature. Pediatrics1999;103(4):E53.

Janssen PA, Selwood BL, Dobson SR, et al. To dye or not to dye: a randomized, clinical trial of a triple dye/alcohol regimen versus dry cord care. Pediatrics 2003;111(1):15–20.

Mason WH, Andrews R, Ross LA, et al. Omphalitis in the newborn infant. Pediatr Infect Dis J 1989;8(8):521–525.

Solomkin J, Mazuski J, Baron E, et al. Guidelines for the selection of anti-infective agents for complicated intra-abdominal infections. Clin Infect Dis 2003;37:997–1005

Vural C, Gungor A, Comerci S. Accuracy of computerized tomography in deep neck infections in the pediatric population. Am J Otolaryngol 2003;24:143–148.

Wong CH, Chang CH, Pasupathy S, et al. Necrotizing fasciitis: clinical presentation microbiology, and determinants of mortality. J Bone Joint Surg Am 2003;85A(8):1454–1460.