Strange and Schafermeyer's Pediatric Emergency Medicine, Fourth Edition (Strange, Pediatric Emergency Medicine), 4th Ed.

CHAPTER 61. Soft Tissue Infections

Coburn H. Allen

Sujit S. Iyer

Donna M. Moro-Sutherland


• Bacterial (septic) arthritis occurs most commonly in children younger than 3 years. Staphylococcus aureus is the most common cause of bacterial arthritis in all age groups.

• Prepatellar bursitis (septic) is seen in children with local cellulitis and often local trauma. Children will present with local signs of infection and preservation of joint function. Treatment is focused on local aspiration and drainage and antibiotics targeted at S. aureus.

• Discitis presents in children most commonly with abnormal gait or lower back pain. Clinical improvement comes with early anti-inflammatory medications and antibiotics targeted at S. aureus and Kingella kingae.

• Clinical manifestation of infectious tenosynovitis ranges from pain with passive extension to tenderness along the tendon sheath. Management includes surgical intervention and antibiotic therapy

• Osteomyelitis typically develops after a period of bacteremia and presents with fever, and progressively increasing bone pain or limp. S. aureus is the most common cause of acute hematogenous osteomyelitis in children; however, K. kingae is increasingly identified in preschool aged children with osteoarticular infections.


Infections of the joint are most commonly bacterial but may be caused by an array of organisms (i.e., fungal or viral). The term septic arthritis encompasses bacterial arthritis, pyogenic arthritis, suppurative arthritis, purulent arthritis, and pyarthrosis. It occurs most commonly in childhood; children younger than 3 years are affected most frequently. Boys are affected more often than girls (male-to-female ratio of 1.2–2:1). Infections of the knee, hip, and ankle account for at least 80% of cases, with the hip and knee most commonly affected. Early diagnosis and treatment of a septic hip is essential in preserving function. Delay in treatment increases the risk of complications, including osteonecrosis of the capital femoral epiphysis, osteomyelitis, chondrolysis, systemic sepsis, and secondary osteoarthritis.1


Normal joints contain a small amount of synovial fluid, which is viscous, clear, and mostly acellular. Microorganisms can enter the joint space by hematogenous spread (which accounts for most cases), direct inoculation, or extension of a contiguous focus of infection, that is osteomyelitis. The high, effective blood flow and lack of basement membrane in the synovium facilitate the entry of bacteria into the joint space during episodes of bacteremia. Host cells responding to bacterial endotoxin release cytokines, which stimulate the release of proteolytic enzymes and increase leukocyte migration, thus destroying the synovium and collagen matrix and inhibiting cartilage synthesis. Gram-positive organisms, specifically S. aureus, are the most common causative organisms (see Table 61-1 for other organisms to consider).

TABLE 61-1

Organisms Most Commonly seen with Septic Arthritis



Septic knee and ankle joints are typically swollen, red, warm, and tender to palpation. Active and passive range of motion is usually decreased and painful. Pain increases with maneuvers that increase intracapsular pressure (e.g., compression of the head of the femur into the acetabulum). When septic arthritis of the hip is suspected there is a history of nontraumatic progressive pain in the hip, limp, and fever and irritability; asymmetric buttock creases or swelling of the buttock or the genitalia may be present with unilateral hip involvement. See Table 61-2 for signs and symptoms.

TABLE 61-2

Signs and Symptoms of Septic Arthritis


Severe symptoms do not always imply severe disease; toxic synovitis can immobilize the hip with pain but is a benign and self-limited condition (see Chapter 110). A child who is unable to bear weight in the ED should not be discharged until a diagnosis is made and therapy instituted.


The laboratory evaluation of a child in whom septic arthritis is suspected includes a complete blood count (CBC) with differential, acute phase reactants (erythrocyte sedimentation rate [ESR] and C-reactive protein [CRP]), and blood culture (BC).2 Arthrocentesis provides synovial fluid for Gram stain, culture (aerobic and anaerobic with susceptibility testing), and white blood cell (WBC) count with differential, and may also relieve pain by decompressing the joint space (see Chapter 108).

For hip involvement, radiographs should be obtained with the child in the “frog leg position” (femur externally rotated, flexed, and abducted)(see Chapter 108). Early radiographic findings of septic arthritis include widening of the joint space, subluxation, and soft-tissue swelling whereas late findings include erosion of the epiphyses and erosion of adjacent subchondral bone (indicative of concurrent osteomyelitis). Ultrasound can be used to guide diagnostic aspiration, and lack of visualized fluid also has a high negative predictive value. The sonographic characteristics of the fluid (i.e., echolucent vs. echogenic) do not correlate with infection, but ultrasound can discern cartilaginous structures. MRI is highly sensitive for the early detection of joint fluid, and can demonstrate abnormalities of adjacent bone and soft tissue, the extent of cartilage destruction, and concomitant osteomyelitis or osteomyelitis-associated abscess.3 It should not be utilized as a first-line diagnostic modality for septic arthritis but is often considered for those who fail to respond after 48 hours of appropriate antimicrobial therapy for bacterial arthritis and initial drainage or aspiration of the affected joint.


Transient/toxic synovitis in contrast to septic arthritis, is self-resolving, managed with supportive therapy, and the most prevalent disease process in the differential diagnosis (see Chapter 100). Four predictors initially described in models to distinguish toxic synovitis from septic arthritis are a history of fever, non–weight-bearing, an ESR of at least 40 mm/h and a WBC count of more than 12,000 cells/mm3.3 In 2006, Caird et al. added CRP >2 mg/dL to the predictive model.4,5 (Table 61-3)27 The predictive model can be applied to a high-risk group with a high prevalence of septic arthritis (i.e., tertiary care center) but its performance may be limited in community-based EDs where the prevalence and presentation is different.

TABLE 61-3

Predictors in Differentiating Septic Arthritis from Toxic Synovitis



Early delivery of antibiotics is widely supported. Microbiological evidence of infection often is not possible and awaiting cultures should not delay treatment. Length of IV antibiotics varies between 2 and 6 weeks with growing evidence in support of shortening such a regimen and converting over to oral antibiotics as signs and symptoms improve (Table 61-4).

TABLE 61-4

Diagnosis and Treatment


Surgical drainage and antimicrobial therapy are the cornerstones of therapy. Treatment includes sterilization and decompression of the joint space and removal of inflammatory debris to relieve pain and prevent deformity or functional sequelae. Drainage can be accomplished through needle aspiration or open surgery (arthrotomy). Delayed drainage increases the likelihood of adverse outcome. Increased intra-articular pressure can compromise the blood supply to the femoral head, resulting in avascular necrosis. Antibiotic therapy (Table 61-5) can be completed on an outpatient basis of 5 to 10 days after initiation of antibiotic therapy in patients with unequivocal clinical improvement (afebrile for 48–72 hours, local signs and symptoms of infection are reduced considerably, WBC normalized, and CRP and/or ESR have decreased).

TABLE 61-5

Intravenous Antibiotic Therapy


Despite antimicrobial treatment, the inflammatory process may be prolonged, leading to delayed recovery and residual joint damage. Randomized trials with dexamethasone as adjunct to antimicrobials for the first 4 days of therapy demonstrated a shorter duration of both local and systemic inflammatory signs, a shorter duration of elevated acute phase reactants, a shorter duration of IV treatment and hospital stay, and a lower risk of residual dysfunction 12 months after treatment (2% vs. 26%).8


Bursae are sack-like structures that protect the bony prominences and provide a cushion for muscle, tendons, and skin to contract and easily slide over the bone. The prepatellar bursa is a superficial bursa that is located between the patella and the skin. Bursitis can occur from repetitive motion, but this review will focus on infectious (septic) bursitis. Those with repeated kneeling and trauma are at higher risk (e.g., young children, athletes, and occupations that involve kneeling such as housekeepers, plumbers, and gardeners).

The prepatellar bursa lies directly under the skin; superficial abrasions and inoculations can easily seed the bursa with bacteria. Surrounding cellulitis is a risk factor. Rarely, prepatellar bursitis may occur from bacteremic spread from an extraneous source. S. aureus is the most common pathogen (>75%) found in culture-proven aspiration. Streptococcus pyogenes is the second most common reported organism in pediatric case series.9,10


The most common symptoms include acute local redness, warmth, and tenderness to the center of the bursa with preservation of joint motion. Pain-free joint motion is important in differentiating this disease from septic arthritis. Fever is present in the majority of patients, and many patients may have concurrent cellulitis or signs of trauma.9,11 Careful attention should be paid to puncture wounds that may be at higher risk for foreign bodies.


Aspiration of fluid is recommended in most cases since a fluctuant pre-patellar region is a common finding. Due to its superficial location, needle aspiration or superficial incision and drainage is often adequate treatment in combination with anti-staphylococcal and streptococcus antimicrobials. Cultures should be obtained of aspirated fluid to direct antibiotic treatment.9,12 More extensive wound exploration is necessary in traumatic cases with possible foreign body retention. BCs are only necessary in toxic appearing patients or cases where there is suspicion for underlying osteomyelitis. IV antibiotics are only necessary in cases of severe local inflammation or an underlying immunocompromised state that could predispose to slow recovery.12 Patients that have persistent systemic symptoms, pain, or develop restrictive joint motion, should be suspected to have underlying osteomyelitis or septic arthritis and may need surgical treatment for diagnosis and cure.


Aspiration of fluid with Gram stain and culture is the most useful diagnostic and therapeutic intervention. White blood counts from aspirated fluid will be lower than found in infected joints.9 Radiographs are useful when there is suspicion for traumatic injury to the bone. Ultrasound can delineate the boundaries of the fluid collection and help with identifying some foreign bodies, but can often not differentiate between infectious and noninfectious bursitis.13MRI should be reserved for cases in which osteomyelitis is of higher suspicion. Peripheral white count, ESR, and CRP are often not needed to dictate treatment for simple prepatellar bursitis.


Tenosynovitis is the inflammation and subsequent fibrosis of the tendon and its enveloping sheath, commonly involving the upper extremity, particularly the hand. Noninfectious causes result from overuse, direct trauma and/or a systemic disease. “Trigger finger” or stenosing flexor tenosynovitis, is an overuse injury most commonly seen in the finger or thumb. Infectious causes result from a pathogen gaining entry to the synovium resulting in an inflammatory process and disrupting the normal tendon function.

Infectious tenosynovitis (less common than noninfectious) develops via: (1) trauma with direct inoculation (i.e., puncture or bite, laceration); (2) contiguous spread from adjacent infected soft tissues; and/or (3) hematogenous spread (Table 61-6).14 When a bacterial infection gains entry to the synovium, it creates an inflammatory process between the inner visceral layer and outer parietal layer of the synovium spreading throughout the entire compartment of the synovium (not simply localized to the point of friction as in the noninfectious types). In flexor tenosynovitis, the tendons are overlaid by retinacular structures that act as pulleys to facilitate movement. With infection, the flexor tendons and synovium are trapped in loculated pus, which can progress rapidly to rupturing and necrosis of the tendon sheath. Compartment syndrome and tissue necrosis can result from spread of infection from tendon sheaths to the surrounding soft tissue. In contrast, most extensor tendons lack retinacular structures making it difficult to distinguish these infections from simple soft-tissue infections.

TABLE 61-6

Mechanism of Infection and Organisms for Infectious Tenosynovitis



In noninfectious tenosynovitis, the pain on motion will be more localized to the site of greatest friction and/or nodules on the tendon may be appreciated on examination.

Flexor tenosynovitis is an infection of the flexor tendon sheath. The four classic signs (referred to as Kanavel’s sign) are: (1) a flexed resting posture, (2) tenderness over the flexor sheath, (3) fusiform swelling, and (4) pain on passive extension. The process has the ability to rapidly destroy a finger’s functional capacity and is considered an orthopedic emergency.14


A CBC with differential, ESR, CRP may help delineate infectious versus noninfectious causes. Obtain BCs with known or suspected pyogenic tenosynovitis and systemic symptoms and/or signs (fever, chills, or hypotension). TB testing should be considered, particularly for the patient with a “boggy” form of tendon nodularity seen with Mycobacterium tuberculosis. If considering juvenile rheumatoid arthritis, an antinuclear antibody (ANA) panel and rheumatoid factor (RF) should be drawn. If drainage is present, specimens should be sent for Gram stain and culture. In most cases, aspiration and/or biopsy is required for both diagnostic and therapeutic purposes.

Plain radiographs will not diagnose tenosynovitis, yet are helpful when considering additional findings such as a foreign body or fracture. Ultrasound may identify the thickening tendon or synovium but its sensitivity and accuracy are technician dependent. The thickened tendon, length affected and synovium can be seen on MRI and is the radiologic study of choice if the clinical findings are equivocal.


Noninfectious tenosynovitis is initially treated with: (1) discontinuation of overuse activity, (2) immobilizing the affected tendon, (3) NSAID, and (4) ice. As the initial discomfort resolves, gradual return of activity and active stretching to reduce long-term stiffness should be performed. Glucocorticoids facilitate breakdown of the thickened tendon and are considered in the patient with persistent stiffness 4 to 6 weeks after initial immobilization. Pediatric “trigger thumb” represents a special population with >60% of the patients demonstrating resolution of symptoms without any treatment.15

Infectious tenosynovitis can be treated medically and/or surgically depending on progression at the time of presentation. See Table 61-714 for selection of antibiotic for initial management. Surgical decompression with clean out should be considered for progressive or moderate cases. As inflammation worsens, there is limited ability of the antibiotic to penetrate the tendon and synovium, thus necessitating surgical interventions. Consultation with a surgeon is important, especially in high-risk areas as the staging of the infection has prognostic implications. (Table 61-8)16

TABLE 61-7

Antibiotic Therapy for Infectious Tenosynovitis


TABLE 61-8

Treatment Based on Michon Classification for Infectious Tenosynovitis


Despite early treatment with appropriate antibiotics, surgical drainage, and irrigation, many patients with infectious tenosynovitis develop decreased range of movement, stiffness, soft-tissue necrosis or worse, amputation of the digit. Animal studies utilizing locally administered corticosteroids in conjunction with antibiotics in pyogenic flexor tenosynovitis demonstrate improved range of motion of the affected tendon. In addition, locally administered antibiotics showed additional benefit for improving range of motion. Future studies will be needed to determine if such therapy will be efficacious in humans.17


Acute hematogenous osteomyelitis (AHO) is the most common type of bone infection in children. “Acute” osteomyelitis has traditionally been defined as having symptoms for less than 2 weeks, while “subacute” and “chronic” osteomyelitis are used to describe children with delayed diagnosis or those who do not fully respond to therapy.18 AHO, like other musculoskeletal infections, is thought to occur when a blood stream infection allows for inoculation of tissue, and in the case of AHO this is usually in the metaphyseal region of long bones where slow blood flow occurs in an area also prone to injury.18Once established, bacteria begin to destroy bone, spread to surrounding tissues, and may become a nidus for persistent bacteremia leading to sepsis and multifocal disease. If not recognized early and aggressively managed AHO may progress to a chronic form characterized by necessitation of prolonged antibiotic courses and surgical debridement of necrotic bone.18

AHO is rare, however, it is more common in children than adults, and has a propensity for boys, recent trauma, and certain comorbidities such as sickle cell anemia (Salmonella sp). K. kingae AHO (and septic arthritis) tends to occur in younger children, particularly those of preschool age in daycare.19 Long bone involvement is most common, with the femur, tibia, and humerus accounting for the majority of cases.20 Interestingly, atypical organisms (Gram negatives, tuberculosis, anaerobes, cat scratch disease and fungi) seem to have predilections for “atypical” bones (i.e., vertebra, pelvis, and other cuboidal, irregular or flat bones) compared to S. aureus.18


The classic presentation of AHO is a child with current or recent fever and progressive limp or extremity pain; however, focal bone tenderness (especially metaphyseal) may be the only finding at presentation. The ability to clinically discriminate the location of pain is diminished in nonverbal and young children. The presence of overlying warmth, redness, and swelling is variable, but range of motion is typically spared unless the adjacent joint is infected also. Most cases involve a single bone; however, multifocal bone disease with increased virulence has been noted in recent years, as have cases involving the adjacent joint and muscle.21,22 However, symmetric bone or primary diaphyseal disease is quite rare with AHO and should alert the clinician to noninfectious processes.

Most children with AHO will have a normal white blood count, about a third to half will have a positive BC and almost all have elevated inflammatory markers in the CA-MRSA era.21,22 Plain films of the involved bone are usually normal (80%) at presentation, but subtle signs of periosteal elevation of surrounding edema may exist.23


AHO may coincide with other musculoskeletal infections (especially myositis, and septic arthritis), but may present similarly to them as well. Children also may have a truly unrelated injury or bone disease (i.e., fracture, avascular necrosis) and a separate infection (viral or bacterial) that will confuse matters. Rheumatologic processes that may mimic AHO include oligoarticular juvenile idiopathic arthritis, rheumatic fever, chronic recurrent multifocal osteomyelitis, and the early stages of Hennoch–Schoenlein Purpura. Bone pain from ischemia in children with hemoglobinopathies can be very difficult to distinguish from AHO.


Guidelines developed by the Infectious Disease Society of America in collaboration with the American Academy of Pediatrics and Pediatric Infectious Disease Society provide recommendations for the management of pediatric bone and joint infections.24

Occasionally a child with AHO will present acutely ill with sepsis, but typically a child will be stable enough to allow clinicians to use a progressive evaluation and management strategy. Given the prolonged nature of antibiotic treatment required for AHO (typically 4–6 weeks for long bones), a focus on maximizing a microbiologic diagnosis has led to recommendations that a brief delay in antibiotics is reasonable while obtaining appropriate blood and tissue cultures and histology in stable children.

The preliminary evaluation of a child with possible AHO in the ED includes obtaining a CBC, ESR, CRP, and one or more BCs. Although typically normal in AHO, plain films of the involved bone (or entire extremity in younger children unable to localize pain) are recommended to rule out noninfectious causes or to confirm a subacute or chronic osteomyelitis. Early consultation of an orthopedic surgeon is helpful, especially in cases where joints may be involved and require aspiration or open drainage, or if a delay in beginning antibiotics cannot be justified secondary to clinical condition.

Increasingly, the confirmation of AHO is made by obtaining an MRI, which has very high sensitivity and specificity for the diagnosis.18,23,24 If the area of involvement is uncertain (e.g., a febrile toddler who will not walk), a bone scan may be helpful in localizing disease.23

In acutely ill or septic, appearing children, broad-spectrum parenteral antibiotics designed to cover the most common pathogens based on age and risk factors should begin as soon as BCs are obtained. Commonly, the administration of antibiotics can be delayed pending urgent obtaining of tissue cultures when feasible. Recent studies in younger children suggesting similar rates of osteoarticular infections caused by K. kingae as with S. aureus have led some experts to recommend adding a third generation cephalosporin to anti-staphylococcal coverage in this population. Antibiotics selected should be informed by local patterns of resistance, particularly when considering rates of MRSA versus MSSA, and clindamycin resistance among all S. aureus strains. Typically, vancomycin, clindamycin, or an anti-staphylococcal penicillin (nafcillin or oxacillin) is given. A third-generation cephalosporin may be added to cover gram-negative organisms in high risk patients (i.e., children with hemoglobinopathies or very young children are at risk for Kingella).

True recurrences of AHO are rare in children and likely are due instead to relapse (chronic osteomyelitis) within the same bone, or unrecognized (and often undrained) multifocal osteomyelitis.

Aside from new antibiotics, advances in management of AHO include a promising diagnostic tool for children with culture-negative osteomyelitis: polymerase chain reaction testing for species-specific ribosomal genes (16s rDNA) 3, 4, 18.25 In addition, progress toward a S. aureus vaccine continues, as more is understood about conserved antigens and normal host response to the bacterium.26


Although earlier literature hypothesized a traumatic or inflammatory etiology for discitis, most current opinion points to the cause as a low-grade infective spondylitis along the spectrum of vertebral osteomyelitis. Due to the mild presentation of clinical symptoms, there are likely many underdiagnosed cases. Case series reveal a peak in toddlers and early to mid-adolescence.2730

Biopsy is not necessary to confirm the diagnosis and BCs are often negative. As such, most case series report no single pathogen as the primary cause. However, in studies where biopsies are performed, S.aureus is the predominant pathogenic organism found followed by K. kingae31. The likelihood of an identifying pathogen increases in patients with concomitant vertebral osteomyelitis. Several studies have proposed that the rich and anastomotic blood supply to the vertebral end plates of the young child allow them to limit the infection to the disc without allowing progression to the bone. This is further supported by the increased rate of bone destruction and osteomyelitis seen in adults.


Young toddlers often present with gait abnormalities, including limp and refusal to walk or crawl. Older children will often complain of back pain and pain with bending forward.27,28 Symptoms can also mimic other pathology including meningitis (irritability), abdominal pain, muscle weakness, and rarely neurological deficits.30 Fever is almost universally absent or low grade.27,28,30 Presentation is subacute with symptoms lasting 2 to 3 weeks before the diagnosis is made.27,28,30 Patients who are ill-appearing or have higher fevers should raise the suspicion of vertebral osteomyelitis.27


Due to the lack of definitive studies identifying a bacterial pathogen and a significant percentage improving with rest and supportive care, there is insufficient research to support a consensus on optimum treatment. As a result, management has focused on a combination of anti-inflammatory medications, immobilization, and antibiotics. Most case series in the last two decades have focused on early antibiotics and anti-inflammatory medications, with some support that earlier IV antibiotics are more likely to lead to rapid relief of symptoms and less chance of recurrence.27 Initial antibiotic treatment is targeted at S. aureus and K. kingae with use of an antistaphylococcal agent and a third generation cephalosporin. Immobilization may be of greatest use in the early phase of the disease for pain control.


Peripheral white counts and ESR tend to be minimally elevated but nonspecific. BCs have a low yield in identifying a pathogen, although they may be useful if suspicion for vertebral osteomyelitis exists. With the high rate of resolution with conservative treatment, biopsy is rarely needed in cases of limited discitis. Plain radiographs may confirm the diagnosis with the right clinical picture. Radiographs can be less reliable early in the disease process, and most centers consider MRI the gold standard for diagnosis.


1. Sultan J, Hughes PJ. Septic Arthritis or transient synovitis of the hip in children: The value of clinical prediction algorithms. J Bone Joint Surg (Am). 2010;92-B(9):1289–1293.

2. Kang SN, Sanghera T, Mangwani J, Paterson JM, Ramachandran M. The management of septic arthritis in children. Systematic Review of the English Language Literature. J Bone Joint Surg (Am).2009;91-B(9):1127–1133.

3. Yang WJ, Im SA, Lim GY, et al. MR imaging of transient synovitis: differentiation from septic arthritis. Pediatr Radiol. 2006;6:1154–1158.

4. Kocher MS, Zurakowski D, Kasser JR. Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. J Bone Joint Surg (Am).1999;81-A(12):1662–1670.

5. Kocher MS, Mandiga R, Murphy JM, et al. A clinical practice for treatment of septic arthritis in children. efficacy in improving process of care and effect on outcome of septic arthritis of the hip. J Bone Joint Surg (Am).2003;85-A(6):994–999.

6. Caird MS, Flynn JM, Leung YL, Millman JE, D’Italia JG, Dormans JP. Factors distinguishing septic arthritis from transient synovitis of the hip in children: a prospective study. J Bone Surg (Am). 2006;88-A:1251–1257.

7. Paakkonen M, Kallio MJT, Kallio PE, Peltola H. Sensitivity of erythrocyte sedimentation rate and C-reactive protein in childhood bone and joint infections. Clin Orthop Relat Res. 2010;468:861–866.

8. Harel L, Prais D, Bar-On E, et al. Dexamethasone therapy for septic arthritis in children. Results of a randomized double-blind placebo-controlled study. J Pediatr Orthop. 2011;31(2):211–215.

9. Harwell JI, Fisher D. Pediatric septic bursitis: case report of retrocalcaneal infection and review of the literature. Clin Infect Dis. 2001;32:E102–E104.

10. Gómez-Rodríguez N, Méndez-García MJ, Ferreiro-Seoane JL, Ibáñez-Ruán J, Penelas-Cortés Bellas Y. [Infectious bursitis: study of 40 cases in the pre-patellar and olecranon regions]. Enferm Infecc Microbiol Clin.1997;15:237–242.

11. Stell IM, Gransden WR. Simple tests for septic bursitis: comparative study. 1998;BMJ. 316:1877.

12. Small LN, Ross JJ. Suppurative tenosynovitis and septic bursitis. Infect Dis Clin North Am. 2005;19:991–1005.

13. Chau CLF, Griffith JF. Musculoskeletal infections: ultrasound appearances. Clin Radiol. 2005;60:149–159.

14. McCarthy JJ, Dormans JP, Kozin SH, Pizzutillo PD. Musculoskeletal infections in children: basic principles and recent advancements. J Bone Joint Surg Am. 2004;86-A(4):850–863.

15. Baek GH, Kim JH, Chung MS, Kang SB, Lee YH, Gong HS. The natural history of pediatric trigger thumb. J Bone Joint Surg Am. 2008;90:980–985.

16. Pang HN, Teoh LC, Yam AKT, Lee JY, Puhaindran ME, Tan AB. Factors affecting the prognosis of pyogenic flexor tenosynovitis. J Bone Joint Surg Am. 2007;89:1742–1748.

17. Draeger RW, Singh B, Bynum DK, Dahners LE. Corticosteroids as an adjunct to antibiotics and surgical drainage for the treatment of pyogenic flexor tenosynovitis. J Bone Joint Surg Am. 2010;92:2653–2662.

18. Krogstad P. Osteomyelitis. In: Feigin RD, Cherry JD, Demmler GJ, Kaplan SL, eds. Feigin and Cherry’s Textbook of Pediatric Infectious Diseases. 6th ed. Vol 2.

19. Ceroni D, Cherkaoui A, Ferey S, Kaelin A, Schrenzel J. Kingella Kingae osteoarticular infections in young children: clinical features and contribution of a new specific real-time PCR assay to the diagnosis. J Pediatr Orthop.2010;30:301–304.

20. Bocchini CE, Hulten KG, Mason EO Jr, Gonzalez BE, Hammerman WA, Kaplan SL. Panton-valentine leukocidin genes are associated with enhanced inflammatory response and local disease in acute hematogenous Staphylococcus aureus osteomyelitis in children. Pediatrics. 2006;117:433–440.

21. Hawkshead JJ, Patel NB, Steele RW, Heinrich SD. Comparative severity of pediatric osteomyelitis attributable to methicillin-resistant versus methicillin-sensitive Staphylococcus aureus. J Pediatr Orthop.2009;29:85–90.

22. Saavedra-Lozano J, Mejias A, Ahmad N, et al. Changing trends in acute osteomyelitis in children: Impact of methicillin-resistant Staphylococcus aureus Infections. J Pediatr Orthop. 2008;(28):569–575.

23. Browne LP, Mason EO, Kaplan SL, Cassady CI, Krishnamurthy R, Guillerman RP. Optimal imaging strategy for community-acquired staphylococcus aureus musculoskeletal infections in children. Pediatr Radiol. 2008;38:841–847.

24. Woods CR, et al. Pediatric bone and joint infection guideline (in development). Infectious Disease Society of America/American Academy of Pediatrics/Pediatric Infectious Diseases Society.

25. Fenollar F, Roux V, Stein A, Drancourt M, Raoult D. Analysis of 525 samples to determine the usefulness of PCR amplification and sequencing of the 16S rRNA gene for diagnosis of bone and joint infections. J Clin Micro.2006;44:1018–1028.

26. McKenna M. Man versus MRSA. Nature. 2012;482:23–25.

27. Fernandez M, Carrol CL, Baker CJ. Discitis and vertebral osteomyelitis in children: an 18-year review. Pediatrics. 2000;105:1299–1304.

28. Brown R, Hussain M, McHugh K, Novelli V, Jones D. Discitis in young children. J Bone Joint Surg Br. 2001;83:106–111.

29. Kayser R, Mahlfeld K, Greulich M, Grasshoff H. Spondylodiscitis in childhood: results of a long-term study. Spine. 2005;30:318–323.

30. Karabouta Z, Bisbinas I, Davidson A, Goldsworthy LL. Discitis in toddlers: a case series and review. Acta Paediatr. 2005;94:1516–1518.

31. Chandrasenan J, Klezl Z, Bommireddy R, Calthorpe D. Spondylodiscitis in children: a retrospective series. J Bone Joint Surg Br. 2011;93:1122–1125.